Provided for non-commercial research and education use. Not for reproduction, distribution or commercial use.

The Canadian Journal of Veterinary Research is published by Canadian Veterinary Medical Association. The attached copy is furnished to readers for personal, non-commercial research and education use. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party Web sites are prohibited. Those who require further information regarding reprints or archiving and manuscript policies are encouraged to contact [email protected]. CanadianCanadian JournalJournal ofof VVeterinaryeterinary RResearchesearch RevueRevue CanadienneCanadienne dede RRechercheecherche VVétérinaireétérinaire

Articles Comparison of levels and duration of detection of antibodies to bovine viral diarrhea virus 1, bovine viral diarrhea virus 2, bovine respiratory syncytial virus, bovine herpesvirus 1, and bovine parainfluenza virus 3 in calves fed maternal colostrum or a colostrum- replacement product Manuel F. Chamorro, Paul H. Walz, Deborah M. Haines, Thomas Passler, Thomas Earleywine, Roberto A. Palomares, Kay P. Riddell, Patricia Galik, Yijing Zhang, M. Daniel Givens ...... 81

Effect of the rider experience and evaluator expertise on subjective grading of lameness in sound and unsound sports horses under saddle Fernando J. Marqués, Cheryl Waldner, Stephen Reed, Fernando Autet, Louise Corbeil, John Campbell ...... 89

Acute exercise does not induce an acute phase response (APR) in Standardbred trotters Lena Kristensen, Rikke Buhl, Katarina Nostell, Lars Bak, Ellen Petersen, Maria Lindholm, Stine Jacobsen ...... 97

Recovery from desflurane anesthesia in horses with and without post-anesthetic xylazine Turi K. Aarnes, Richard M. Bednarski, Alicia L. Bertone, John A.E. Hubbell, Phillip Lerche ...... 103

Les biofilms bactériens : leur importance en santé animale et en santé publique Yannick D.N. Tremblay, Skander Hathroubi, Mario Jacques ...... 110

Immunogenicity and efficacy of a recombinant adenovirus expressing hemagglutinin from the H5N1 subtype of swine influenza virus in mice Yunpu Wu, Chuanling Qiao, Huanliang Yang, Yan Chen, Xiaoguang Xin, Hualan Chen . . 117

Porcine salivary analysis by 2-dimensional gel electrophoresis in 3 models of acute stress: A pilot study María Fuentes-Rubio, José J Cerón, Carlos de Torre, Damián Escribano, Ana M. Gutiérrez, Fernando Tecles . . . . 127

April/Avril 2014, Vol. 78 No. 2 table of contents continued on back cover/ la table des matières se poursuit sur la couverture arrière FOR PERSONAL USE ONLY Established in 1937 ISSN 0830-9000

Canadian Journal of Revue canadienne de Veterinary Research recherche vétérinaire Formerly CANADIAN JOURNAL OF COMPARATIVE MEDICINE Remplace la REVUE CANADIENNE DE MÉDECINE COMPARÉE The journal publishes the results of original research in veterinary La revue publie les résultats de recherches originales en médecine and comparative medicine. Manuscripts must be as concise as pos- vétérinaire et comparative. Les manuscrits doivent être aussi brefs que sible, and the research described must represent a significant con- possible et la recherche décrite doit apporter une contribution impor- tribution to knowledge in veterinary medicine. Full-length papers, tante à la médecine vétérinaire. Les exposés complets, les communi­ short communications, and review papers are welcome. All manu- cations brèves et les comptes rendus sont bienvenus. Nous vérifierons­ scripts will be reviewed for scientific content and editorial accuracy. le fonds ­scientifique et la forme de tous les manuscrits. Manuscripts must conform to the Instructions for Authors found Les manuscrits doivent être conformes aux directives aux auteurs on the CVMA website. Consultation of a recent issue of the journal qui se trouvent sur le site Web de l’ACMV. La consultation d’un to ensure that the manuscript conforms to current style is also recom- récent numéro de la revue est aussi profitable, afin de s’assurer qu’un mended. Please use the website (www.canadianveterinarians.net) manuscrit se conforme au style courant. Veuillez utiliser le site Web to submit articles. A publication charge of $90 per published page, (www.veterinairesaucanada.net) pour soumettre les articles. Les plus a cost of $10 per table, $15 per illustration, and $150 per color auteurs doivent acquitter des frais de 90 $ par page publiée, 10 $ illustration is payable by contributors. Instructions for ordering pour chaque tableau, 15 $ pour chaque illustration, et 150 $ pour reprints are sent with the galley proofs. Color illustrations will only chaque illustration en couleur. Un formulaire de commande de be published if the author(s) agree to pay the extra cost. tirés à part accompagne l’en­voi des épreuves d’un manuscrit. Les illustrations en couleurs ne seront publiées que si les auteurs en assument les frais supplémentaires.

Canadian Journal of Veterinary Research Abstracted and/or indexed in:/Résumé et/ou catalogué dans : Revue canadienne de recherche vétérinaire Biological Abstracts Index Veterinarius Editor — Rédacteur Chemical Abstracts Nutrition Abstracts Éva Nagy, Guelph, Ontario Current Contents (Agric. Section) Science Citation Index Associate Editor — Rédacteur adjoint EMBASE/Excerpta Medica Veterinary Bulletin Deborah M. Haines, Saskatoon, Saskatchewan Index Medicus Derwent Veterinary Assistant Editors — Assistants à la rédaction Drug File Serge Messier, Saint-Hyacinthe (Québec) Jeffrey J. Wichtel, Charlottetown, Prince Edward Island Managing Editor — Directrice de la rédaction Heather Broughton, Ottawa, Ontario Assistant Managing Editor — Directrice adjointe à la ­rédaction Stella Wheatley, Ottawa, Ontario Advertising Manager — Gérante de la publicité Laima Laffitte, Wendover, Ontario Editorial Coordinator/Coordonnatrice de la rédaction Kelly Gray, Ottawa, Ontario

Address all correspondence to/Adresser toute correspondance à : Canadian Journal of Veterinary Research Typesetting/Typographie Revue canadienne de recherche vétérinaire AN Design Communications 339, rue Booth Street, Ottawa, Ontario K1R 7K1 www.an-design.ca Tel./Tél. : (613) 236-1162 — Fax/Télécopieur : (613) 236-9681 e-mail/Courriel : [email protected] © Canadian Veterinary Medical Association 2014 L’Association canadienne des médecins vétérinaires 2014

Subscriptions/Abonnements — 2014 Published quarterly by the Canadian Veterinary Medical Association Canada $155 $ ( applicable GST/HST; Publication trimestrielle de l’Association canadienne des TPS/TVH en vigueur) médecins vétérinaires Foreign/Étranger $170 $ US/É.-U.

Return undeliverable Canadian addresses to: 339 rue Booth Street, Ottawa, Ontario K1R 7K1, e-mail: [email protected] FOR PERSONAL USE ONLY Article

Comparison of levels and duration of detection of antibodies to bovine viral diarrhea virus 1, bovine viral diarrhea virus 2, bovine respiratory syncytial virus, bovine herpesvirus 1, and bovine parainfluenza virus 3 in calves fed maternal colostrum or a colostrum-replacement product Manuel F. Chamorro, Paul H. Walz, Deborah M. Haines, Thomas Passler, Thomas Earleywine, Roberto A. Palomares, Kay P. Riddell, Patricia Galik, Yijing Zhang, M. Daniel Givens

Abstract Colostrum-replacement products are an alternative to provide passive immunity to neonatal calves; however, their ability to provide adequate levels of antibodies recognizing respiratory viruses has not been described. The objective of this study was to compare the serum levels of IgG at 2 d of age and the duration of detection of antibodies to bovine viral diarrhea virus 1 (BVDV-1), bovine viral diarrhea virus 2 (BVDV-2), bovine respiratory syncytial virus (BRSV), bovine herpesvirus 1 (BHV-1), and bovine parainfluenza virus 3 (BPIV-3) in calves fed maternal colostrum (MC) or a colostrum replacement (CR) at birth. Forty newborn male Holstein calves were assigned to the CR or the MC group. Group CR (n = 20) received 2 packets of colostrum replacement (100 g of IgG per 470-g packet), while group MC (n = 20) received 3.8 L of maternal colostrum. Blood samples for detection of IgG and virus antibodies were collected from each calf at birth, at 2 and 7 d, and monthly until the calves became seronegative. Calves in the MC group had greater IgG concentrations at 2 d of age. The apparent efficiency of absorption of IgG was greater in the MC group than in the CR group, although the difference was not significant. Calves in the CR group had greater concentrations of BVDV neutralizing antibodies during the first 4 mo of life. The levels of antibodies to BRSV, BHV-1, and BPIV-3 were similar in the 2 groups. The mean time to seronegativity was similar for each virus in the 2 groups; however, greater variation was observed in the antibody levels and in the duration of detection of immunity in the MC group than in the CR group. Thus, the CR product provided calves with more uniform levels and duration of antibodies to common bovine respiratory viruses.

Résumé Les produits de remplacement du colostrum sont une alternative pour fournir une immunité passive aux veaux nouveau-nés; toutefois, leur capacité à fournir des niveaux adéquats d’anticorps reconnaissant les virus respiratoires n’a pas été décrite. L’objectif de la présente étude était de comparer les niveaux d’IgG sériques à 2 jours d’âge et la durée de détection des anticorps contre le virus de la diarrhée virale bovine de type 1 (BVDV-1), le virus de la diarrhée virale bovine de type 2 (BVDV-2), le virus respiratoire syncitial bovin (BRSV), l’herpesvirus bovin de type 1 (BHV-1), et le virus parainfluenza bovin de type 3 (BPIV-3) chez des veaux nourris avec du colostrum maternel (MC) ou du colostrum de remplacement (CR) à la naissance. Quarante veaux nouveau-nés mâles de race Holstein ont été assignés soit au groupe CR ou MC. Les animaux du groupe CR (n = 20) ont reçu deux paquets de substitut de colostrum (100 g d’IgG par paquet de 470 g), alors que les animaux du groupe MC (n = 20) ont reçu 3,8 L de colostrum maternel. Des échantillons sanguins pour la détection d’IgG et d’anticorps contre les virus ont été prélevés de chaque veau à la naissance, à 2 et 7 j d’âge, et à chaque mois jusqu’à ce que les veaux deviennent séronégatifs. Les veaux dans le groupe MC avaient des concentrations d’IgG plus élevées à 2 j d’âge. L’efficacité d’absorption apparente d’IgG était plus grande dans le groupe MC que dans le groupe CR, bien que la différence ne fût pas significative. Les veaux dans le groupe CR avaient des concentrations plus élevées d’anticorps neutralisants envers BVDV durant les 4 premiers mois de vie. Les niveaux d’anticorps contre BRSV, BHV-1, et BPIV-3 étaient similaires dans les deux groupes. Le temps moyen pour atteindre la séronégativité était similaire pour chaque virus dans les deux groupes; toutefois, de plus grandes variations étaient observées dans les niveaux d’anticorps et la durée de détection de l’immunité dans le groupe MC comparativement au groupe CR. Ainsi, le produit CR a fourni des veaux avec des niveaux d’anticorps contre les virus respiratoires bovins communs plus uniformes et de plus longue durée. (Traduit par Docteur Serge Messier)

Department of Clinical Sciences, College of Veterinary Medicine, Auburn University, 1500 Wire Road, Auburn, Alabama 36849, USA (Chamorro, Walz, Passler, Palomares); Department of Pathobiology, College of Veterinary Medicine, Auburn University, 1500 Wire Road, Auburn, Alabama 36849, USA (Walz, Passler, Givens); Department of Veterinary Microbiology, Western College of Veterinary Medicine, 52 Campus Drive, Saskatoon, Saskatchewan S7N 5B4 (Haines); Animal Health Research, College of Veterinary Medicine, Auburn University, 1500 Wire Road, Auburn, Alabama 36849, USA (Riddell, Galik, Zhang); Nutritional Services, Land O’Lakes Animal Milk Products, 1080 County Road F West, Shoreview, Minnesota 55126, USA (Earleywine). Address all correspondence to Dr. Paul H. Walz; telephone: (334) 844-4951; fax: (334) 844-4368; e-mail: [email protected] Received November 29, 2012. Accepted April 2, 2013.

2014;78:81–88 The Canadian Journal of Veterinary Research 81 FOR PERSONAL USE ONLY

Introduction cows with a killed-virus vaccine that included the same antigens. One dairy farm did not vaccinate cattle against these viruses. The Bovine respiratory disease (BRD) is one of the most common calves were removed from their dams immediately after birth and diseases affecting cattle in the United States and is an important strategically assigned to a colostrum-replacement (CR) group or a cause of economic losses in cattle operations worldwide (1,2). maternal-colostrum (MC) group. All 16 calves born on the farm that Viral respiratory pathogens such as bovine viral diarrhea virus 1 did not vaccinate against BVDV-1, BVDV-2, BRSV, BHV-1, and BPIV-3 (BVDV-1), bovine viral diarrhea virus 2 (BVDV-2), bovine respira- were assigned to the CR group. Strategic assignment was necessary tory syncytial virus (BRSV), bovine herpesvirus 1 (BHV-1), and to ensure the presence of virus antibody in the serum of calves from bovine parainfluenza virus 3 (BPIV-3) play an important role in the the nonvaccinated herd after colostrum intake. Four calves from the pathogenesis of BRD because of their ability to impair the integrity vaccinated herds were randomly assigned to the CR group and 20 of the upper respiratory tract, cause immunosuppression, promote were assigned to the MC group. secondary bacterial infection, and cause acute clinical disease (3–6). Group CR received 2 packets of a colostrum-replacement product Antibodies against respiratory viruses transmitted through maternal (Land O’Lakes Animal Milk Products, Shoreview, Minnesota, USA) colostrum protect calves against acute BRD (4,7,8), and calves with providing 100 g of IgG per packet. Group MC received 3.8 L of fresh high serum titers of antibodies against bovine viral diarrhea viruses, or frozen maternal colostrum. The colostrum-replacement product BRSV, BHV-1, and BPIV-3 on arrival at the feedlot have a lower risk was not reconstituted according to the label directions; instead, a of BRD as evidenced by lower morbidity rates and reduced fre- larger water volume, 1.5 L, was used to reconstitute each packet quency of treatment (8,9). As maternally derived immunity decays, so that the volumes of colostrum given to all the calves were the calves become susceptible to acute bovine viral diarrhea (10). The same. The maternal colostrum was evaluated with a colostrometer duration of maternally derived immunity to respiratory viruses in (Biogenics, Mapleton, Oregon, USA). If the specific gravity was 1.055 the calf is directly proportional to the serum level of antibodies to or greater the calf received the maternal colostrum from the dam. BVDV-1, BVDV-2, BRSV, BHV-1, and BPIV-3 ingested and absorbed If the specific gravity was less than 1.055 the calf received stored from colostrum (11,12): the higher the levels, the longer the duration frozen colostrum collected from cows of the same herd that had a (11). However, previous research has shown a large range of initial higher specific gravity. Nine calves in the MC group received stored serum antibody titers in calves that receive maternal colostrum frozen colostrum. Maternal colostrum or colostrum replacement was (13,14) and, as a consequence, variability in the age at which calves administered in a single feeding by bottle within the first 2 h of life. become susceptible to acute BVDV-1, BVDV-2, BRSV, BHV-1, and If the calf did not take the bottle during the first 10 min of feeding, BPIV-3 infections (11,13). an esophageal feeder was used. Seven calves in the CR group and Provision of high-quality colostrum to neonatal calves in a timely 11 calves in the MC group required an esophageal feeder. manner is a key management practice in preventing the failure of After the colostrum feeding, all the calves were moved to indi- passive transfer of immunoglobulins (FPT); however, maternal vidual pens in a calf isolation barn at the Animal Health Research colostrum may not be consistently available. As an alternative, facilities at Auburn University, Auburn, Alabama, USA. This barn commercial colostrum-replacement products may provide ade- provided biosecurity to prevent exposure to BVDV-1, BVDV-2, BRSV, quate passive immunity while reducing the risk of infection with BHV-1, and BPIV-3 during the trial. After 12 h of life the calves were colostrum-transmitted pathogens (15–17). When such a product examined and fed 1.8 L of a commercial milk replacer (Land O’Lakes is used, a minimum of 100 g of IgG is recommended to achieve Animal Milk Products) twice daily until weaned, between 6 and acceptable levels of serum immunoglobulins, and a greater mass 8 wk of age. A calf starter, hay, and free-choice water were offered of IgG (150 to 200 g) increases passive immunity and more reliably to all the calves from 1 wk of age until weaning. Once weaned, all prevents FPT in calves (15,16). The ability of colostrum-replacement the calves were moved to a biosecure paddock at the Animal Health products to provide calves with adequate levels of antibodies against Research facilities, where they had access to free-choice hay, water, BVDV-1, BVDV-2, BRSV, BHV-1, and BPIV-3 and the duration of and a calf-grower grain mix. During the study period the calves were the replacement-derived immunity have not been described. The evaluated once daily and were cared for under the guidelines of objective of this study was to compare the serum levels and dura- Auburn University’s Institutional Animal Care and Use Committee tion of detection of antibodies to BVDV-1, BVDV-2, BRSV, BHV-1, (PRN 2009-1647). and BPIV-3 in neonatal calves that received maternal colostrum or a colostrum-replacement product at birth. Sample testing Blood samples were collected from each calf before the colos- Materials and methods trum feeding for isolation of BVDV-1 and BVDV-2 and testing for antibodies to BVDV-1, BVDV-2, BRSV, BHV-1, and BPIV-3. Serum titers of antibodies against BVDV-1 and BVDV-2 were measured Calves and experimental treatment by virus neutralization (VN), whereas titers of antibodies against Forty newborn male Holstein calves from 3 dairy farms local to BRSV, BHV-1, and BPIV-3 were measured by indirect enzyme-linked Auburn, Alabama, USA were used in this study. Pertinent manage- immunosorbent assay (ELISA). Any calf with positive results was to ment practices at 2 of the dairies included prebreeding vaccination be removed from the study. Additionally, the total serum IgG con- of heifers with a modified live-virus vaccine that included BVDV-1, centration was measured by single radial immunodiffusion in the BVDV-2, BRSV, BHV-1, and BPIV-3 antigens and vaccination of adult serum from these samples and from those obtained when the calves

82 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

10

9 10 8 9 7 8

6 7

5 6

VN antibody titer 5 2 4 VN antibody titer

2 4 3 3 2 Mean log 2 Mean log 1 1 0 0 0 2 d 7 d 1 mo 2 mo 3 mo 4 mo 5 mo 6 mo 7 mo 8 mo 0 2 d 7 d 1 mo 2 mo 3 mo 4 mo 5 mo 6 mo 7 mo 8 mo

Figure 1. Titers of virus-neutralizing (VN) antibody [mean 6 standard Figure 2. Titers of VN antibody against bovine viral diarrhea virus 2 error (SEM)] against bovine viral diarrhea virus 1 (BVDV-1) in calves (BVDV-2) in the same groups of calves. The mean log2 titers were fed a colostrum replacement (CR; darker grey) or maternal colostrum significantly higher in the CR group than in the MC group at 2 d

(MC) within the first 2 h of life. The mean log2 titers were significantly (P = 0.0119), 7 d (P = 0.0074), 1 mo (P = 0.05), 2 mo (P = 0.0024), higher in the CR group than in the MC group at 2 d (P = 0.0194), 7 d 3 mo (P = 0.0098), and 4 mo of age (P = 0.0107). The decay rate was (P = 0.05), and 3 mo of age (P = 0.0427). The decay rate was similar similar in the 2 groups. in the 2 groups. were 2 d old. Failure of passive transfer of immunoglobulins was Indirect ELISA — This assay for antibodies to BRSV, BHV-1, and defined as a total serum IgG concentration of less than 10 g/L at 2 d BPIV-3 was done as previously described (24,25). Briefly, 96-well of age (18). Samples of the maternal colostrum and the colostrum- Immulon 4HBX microtiter plates were coated with BRSV, BHV-1, and replacement product were evaluated for IgG concentration, and BPIV-3 antigen diluted in carbonate buffer (pH 9.6) in alternating the apparent efficiency of absorption (AEA) of IgG was calculated rows with uninfected cell lysate as a control and incubated overnight with the following formula: serum IgG concentration at 2 d of age at 4°C, then the plates were washed and blocked with 0.2% gelatin (g/L) 3 plasma volume (L) 4 total IgG intake (g) (19,20). The plasma (Sigma Chemical Company, St. Louis, Missouri, USA) in carbonate volume was estimated as 9.9% of the body weight. Blood samples buffer. Serum was added to the wells at a dilution of 1:50 (BRSV) were collected from all the calves for virus antibody testing at 2 and or 1:200 (BHV-1 and BPIV-3), and the plates were incubated at 37°C 7 d of age and monthly until the calves became seronegative for each for 1 h. Positive and negative control serum was included in each virus or until 8 mo of age. plate. The plates were washed, and then a 1:5000 dilution of protein Virus neutralization testing — The standard VN microtiter assay G conjugated with horseradish peroxidase (Zymed, San Francisco, was used to detect serum antibodies against BVDV-1 and BVDV-2 California, USA) was added. Single-component 2,29-azino-di(3-ethyl- (21). The BVDV-1 cytopathic strain NADL and the BVDV-2 cyto- benzthiazoline-6-sulfonate (Kirkegaard & Perry Laboratories, pathic strain 125c were used (21). Briefly, after heat inactivation Gaithersburg, Maryland, USA) was used as the enzyme substrate. at 56°C for 30 min, serial 2-fold dilutions (1:2 to 1:4096) of serum Sample and antibody dilutions were made in ELISA working buf- were made in 50 μL of culture medium. For each dilution, 3 wells fer (0.01 M phosphate buffer, pH 7.2, with 0.75 M NaCl and 0.05% of a 96-well polystyrene microtiter plate (Immulon 4HBX; Thermo Tween 20 with 0.2% gelatin). Standard positive-control serum Electron Corporation, Milford, Massachusetts, USA) were inocu- samples were obtained from cattle with high titers of neutralizing lated with an equal volume (50 μL) of culture medium containing antibody against BHV-1 and BPIV-3. Convalescent serum from an a median tissue culture infective dose (TCID50) of the test strain of unvaccinated calf with naturally occurring BRSV infection was used 100 to 500 per milliliter. After inoculation, the plate was incubated at as a positive control for BRSV. Fetal bovine serum (FBS) was used

38.5°C in a humidified atmosphere of 5% CO2 and air for 1 h. Then as a negative control for all the serum samples. All ELISA results 2.5 3 103 Madin–Darby bovine kidney (MDBK) cells in 50 μL of were calibrated against the results for their respective standard culture medium were added to each well. The plate was incubated positive controls and the negative control to give uniformity to the for 72 h and evaluated visually for a cytopathic effect (21,22). Mean results over the time that the tests were done. Optical density (OD) log2 antibody titers were calculated from the endpoint titers for the values were measured with a Benchmark microplate reader (Bio-Rad animals in each group. Seronegativity to BVDV-1 and BVDV-2 was Laboratories, Mississauga, Ontario) at 492 nm and converted to defined as a serum antibody titer less than 2, which equates to a log2 ELISA units (EU) with a software program (Microplate Manager, antibody titer of 0. The mean time to reach seronegative status with version 5.0.1; Bio-Rad Laboratories). The net OD value of the respect to the 2 viruses and the proportion of calves with a serum sample was calculated by subtracting the mean sample OD values titer # 1:16 or a mean log2 antibody titer # 4 per time point were of the wells coated with the cell control from the mean sample OD calculated and compared between the groups. These levels have values of the wells coated with BRSV, BHV-1, and BPIV-3 antigen. been associated with an increased susceptibility to acute BVDV The net OD values were similarly calculated for the positive-control infection (23). and negative-control serum samples (24). Final serum results were

2000;64:0–00 The Canadian Journal of Veterinary Research 83 FOR PERSONAL USE ONLY

60

90 50 80 40 70 60 30 50 40 20 30 Mean antibody level (EU) 20 10 Mean antibody level (EU) 10

0 0 0 2 d 7 d 1 mo 2 mo 3 mo 4 mo 5 mo 6 mo 7 mo 8 mo 0 2 d 7 d 1 mo 2 mo 3 mo 4 mo 5 mo 6 mo 7 mo 8 mo

Figure 3. Levels of antibody against bovine respiratory syncytial virus Figure 4. Levels of antibody against bovine herpesvirus 1 in the same in the same groups of calves, as measured by enzyme-linked immuno- groups of calves, as measured by ELISA. The mean level was significantly sorbent assay (ELISA). The mean level, in ELISA units (EU), 6 SEM was higher in the MC group than in the CR group at 8 mo of age (P = 0.032). significantly higher in the CR group than in the MC group at 7 days of The decay rate was similar in the 2 groups. age (P = 0.0384). The decay rate was similar in the 2 groups.

100 90 100 80 90 70 80 60 70 60 50 50 40 40 % of calves 30

% of calves 30 20 20 10 10 0 0 2 d 7 d 1 mo 2 mo 3 mo 4 mo 5 mo 6 mo 7 mo 8 mo 2 d 7 d 1 mo 2 mo 3 mo 4 mo 5 mo 6 mo 7 mo 8 mo

Figure 5. Proportions of calves with a serum titer # 1:16 or a mean log2 Figure 6. Proportions of calves with a serum titer # 1:16 or a mean log2 titer # 4 of antibody against BVDV-1 in the same groups of calves. These titer # 4 of antibody against BVDV-2 in the same groups of calves. These antibody levels started to be reached earlier and in greater proportions of antibody levels started to be reached earlier and in greater proportions of the calves per time period in the MC group (lighter grey bars) compared the calves per time period in the MC group (lighter grey bars) compared with the CR group; the differences between the groups were significant with the CR group; the differences between the groups were significant only at 2 and 3 mo of age. at 7 d, 1 mo, 2 mo, 3 mo, 4 mo, and 5 mo of age. expressed in EU as follows: (mean net sample OD — mean net reference serum from the Center for Veterinary Biologics, Animal negative-control OD) 4 (mean net positive-control OD — mean and Plant Health Inspection Service, United States Department of net negative-control OD) 3 100. The results for each virus were Agriculture, Ames, Iowa. All samples were tested in triplicate and classified as positive or negative according to whether the ELISA incubated in a humid atmosphere at 25°C for 18 to 24 h. Ring diam- reactivity was greater or less than 10 EU, which is equivalent to a eters were measured with a computer-assisted plate reader (The serum neutralizing-antibody titer of 1:3 for the 3 viruses (26). The Binding Site Group, Birmingham, England) and the values for the mean time to reach seronegative status with respect to the 3 viruses samples calculated with a program for linear analysis. and the proportion of seronegative calves per time period were calculated and compared between the groups. Statistical analysis Single radial immunodiffusion — Undiluted serum samples obtained All statistics were calculated with SAS software (SAS Institute, before colostrum feeding, serum samples diluted 1:4 obtained Cary, North Carolina, USA) with an a level of 0.05. To detect the after colostrum feeding, maternal colostrum diluted 1:15, and decay in serum antibody titer over time after the intake of maternal colostrum-replacement samples diluted 1:15 were assayed for total colostrum or colostrum replacement, the mean titer of antibody IgG concentration by single radial immunodiffusion, essentially as against each virus measured at 2 d of life was compared with the previously described (20). Antiserum against bovine IgG (heavy and mean titer at 7 d and at 1, 2, 3, 4, 5, 6, 7, and 8 mo of life by repeated- light chains; Jackson Laboratories, West Grove, Pennsylvania, USA) measures analysis in a mixed generalized linear model. Analysis of was used. Immunodiffusion plates were prepared from 2% agarose variance (ANOVA) was done to compare the mean time to reach containing 2.5% antiserum in phosphate-buffered saline, pH 7.25. seronegative status among treatment groups with the use of a gen- Standard curves (1.06 to 8.5 g/L) were produced by means of dupli- eralized linear model. Since the antibody data did not have a normal cate samples of a bovine IgG serum calibrator (Midland BioProducts distribution and the variances were not homogeneous between the Corporation, Boone, Iowa, USA). Validity was assessed with a groups, a nonparametric 1-way ANOVA (Kruskal–Wallis test) was

84 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

Table I. Variation in levels of antibody against respiratory viruses and their respective coefficients of variation in groups of calves fed a colostrum-replacement product or maternal colostrum within the first 2 h of life Treatment; measures of variation Colostrum replacement Maternal colostrum Virus Time pointa Extremesb CVc (%) Extremesb CVc (%) P-valued BVDV-1 2 d 8, 10 9.55 5, 11 28.03 0.0084 2 mo 6, 9 12.55 1, 11 51.11 0.0003 5 mo 0, 5 55.52 0, 5 87.92 0.0611 BVDV-2 2 d 7, 10 11.12 2, 11 37.40 0.0028 2 mo 5, 8 13.75 0, 7 57.14 0.0003 5 mo 0, 4 74.21 0, 4 . 100 0.6535 BRSV 2 d 74, 84 3.50 26, 119 24.98 0.0269 2 mo 45, 70 17.59 3, 97 45.32 0.0118 5 mo 0, 21 78.46 0, 44 . 100 0.1802 BHV-1 2 d 33, 71 17.59 11, 85 43.49 0.0002 2 mo 8, 41 44.42 0, 52 61.48 0.0069 5 mo 0, 3 . 100 0, 9 . 100 0.8752 BPIV-3 2 d 83, 187 20.06 22, 168 34.19 0.3416 2 mo 28, 134 35.72 4, 128 55.88 0.0178 5 mo 0, 23 . 100 0, 37 . 100 0.2659 a The 2-d time point represents the time the first sample was obtained after administration of the colostrum-replacement product or maternal colostrum. b Extremes of log2 antibody titer for bovine viral diarrhea virus 1 (BVDV-1) and bovine viral diarrhea virus 2 (BVDV-2) and of enzyme-linked immunosorbent­ assay units for bovine respiratory syncytial virus (BRSV), bovine herpesvirus 1 (BHV-1), and bovine parainfluenza virus 3 (BPIV-3). c Coefficient of variation (CV = s/m 3 100) of the mean levels of antibody against each virus at each time point. d Based on the Levene test for homogeneity of variances. done to compare the mean serum antibody levels for each virus fed with a bottle and those fed with an esophageal feeder within at each time point and to compare the serum and colostrum or the 2 groups (P = 0.098). The AEA at 2 d of age was higher, but not replacement IgG concentrations among the treatment groups. To significantly so (P = 0.2238), in the MC group than in the CR group, compare variation in antibody values between the groups, extremes at 27.21% versus 23.40%. of antibody titers were determined and coefficients of variation (CV) All the calves were seronegative for all the test viruses before calculated at 2 d, 2 mo, and 5 mo of age. The Levene test was used to colostrum intake. The CR group had mean BVDV-1 and BVDV-2 statistically compare the extent of variation of antibody levels within antibody titers that were significantly greater than those in the MC the MC and CR groups. Logistic regression (PROC LOGISTIC) was group at 2 d, 7 d, and 3 mo of age (Figure 1) and at 2 d, 7 d, 1 mo, used to compare the proportion of calves reaching seronegative 2 mo, 3 mo, and 4 mo of age (Figure 2), respectively. For BRSV the status with respect to each virus at each time point between the mean antibody level was significantly higher at 7 d of age in the treatment groups, and significant differences were determined by CR group than in the MC group (81.3 versus 70.65 EU; P = 0.0384); the chi-squared test. however, significant differences were not observed at the other time points (Figure 3). The mean BHV-1 antibody level was signifi- Results cantly higher at 8 mo of age in the MC group than in the CR group (0.47 versus 0.0 EU; P = 0.0328); however, additional significant Before the calves were fed maternal colostrum or the colostrum- differences were not observed (Figure 4). For BPIV-3, statistically replacement product, IgG was not detectable in the serum. The mean significant differences were not observed between the groups at any total mass of IgG administered by feeding was higher with maternal time point. colostrum than with colostrum replacement, at 277.4 versus 208.5 g Table I shows the variation in virus-antibody levels between the (P = 0.0089). The calves in the MC group had higher mean concen- groups at 2 d, 2 mo, and 5 mo of age. For BVDV-1, BVDV-2, BRSV, trations of serum IgG at 2 d of age than the calves in the CR group: and BHV-1 the variation was significantly greater (P , 0.05) in the 20.65 versus 12.41 g/L (P = 0.021). There were no significant differ- MC group than in the CR group at 2 d and 2 mo of age, whereas ences in serum IgG concentration at 2 d of age between the calves for BPIV-3 the variation was significantly greater (P = 0.0178) in the

2000;64:0–00 The Canadian Journal of Veterinary Research 85 FOR PERSONAL USE ONLY

Table II. Mean time for the 2 groups of calves to reach more uniform given that the product is created from pools of large seronegative status with respect to BVDV-1, BVDV-2, BRSV, numbers of individual collections of colostrum that are selected BHV-1, and BPIV-3 according to IgG concentration. However, the CR group received twice the amount of colostrum-replacement product recommended Treatment; mean time 6 standard deviation (mo) by the manufacturer, which may suggest that a single dose of the Virus Colostrum replacement Maternal colostrum product could potentially be inadequate to provide calves with BVDV-1 6.1 6 0.7 5.5 6 1.7 antibody levels similar to those found in this study. BVDV-2 6.5 6 0.6 6.1 6 1.6 There was greater variability in the levels of antibody against BRSV 5.5 6 0.6 5.2 6 1.5 BVDV-1, BVDV-2, BRSV, BHV-1, and BPIV-3 among the calves in BHV-1 3.75 6 0.8 3.8 6 1.6 the MC group compared with the calves in the CR group during BPIV-3 5.1 6 0.8 4.9 6 1.66 the first 2 mo of life. A wide range of levels of antibodies to respira- tory viruses has been reported in calves after maternal colostrum MC group than in the CR group only at 2 mo of age. At 5 mo of age intake (12,13). These levels have a direct effect on the duration of there were no significant differences in variation between the groups. colostrum-derived immunity, as calves with low initial levels become The duration of passively acquired immunity, expressed as the seronegative earlier in life (11,12). Factors such as diversity of mater- mean time to reach seronegative status with respect to BVDV-1, nal colostrum sources, differences in concentration of virus-specific BVDV-2, BRSV, BHV-1, and BPIV-3, did not differ significantly antibodies in maternal colostrum, and prevalence of FPT influence between the CR and MC groups (Table II). Significant decay of the levels of colostrum-derived antibodies in calves fed maternal antibody titers was observed after 1 mo of age for each virus in both colostrum. However, the mean serum IgG concentrations at 2 d of groups, and the decay rate was similar in the 2 groups. The calves life in both groups in this study were above the reported concentra- in the MC group reached serum titers of antibody against BVDV-1 tion associated with FPT in calves (, 10 g/L) (18); therefore, FPT and BVDV-2 of # 1:16 (log2 # 4) in greater proportions at each time was unlikely as a factor contributing to the variability in antibody point compared with the calves in the CR group (Figures 5 and 6, levels observed in the MC group. The lower AEA observed in the CR respectively). However, significant differences between the groups group compared with the MC group and compared with results of were observed only at 2 and 3 mo of age for BVDV-1 and at 7 d, the use of this product in previous studies (15,16) could have been 1 mo, 2 mo, 3 mo, 4 mo, and 5 mo of age for BVDV-2 (P , 0.05). due to the use of more water when reconstituting the product than Similarly, for BRSV and BHV-1 the calves in the MC group started the manufacturer recommended. to become seronegative in greater proportions at each time point Several studies have reported the duration of detection of mater- compared with the CR calves; however, significant differences were nal colostrum-derived antibodies against BVDV-1, BVDV-2, BRSV, observed only at 4 mo of age (P , 0.05). For BPIV-3 the proportion BHV-1, and BPIV-3 as the mean time to reach seronegativity for of calves becoming seronegative at each time point did not differ each virus (11,13). The longevity of passively acquired immunity significantly between the groups. in calves that receive maternal colostrum at birth is highly vari- able (12,14,26,27). One study estimated the mean time to reach Discussion seronegativity for respiratory viruses in a group of calves that had received maternal colostrum as being 117.7 6 37.7 d for BVDV-1, This is the first report demonstrating the efficacy of a colostrum- 93.9 6 61.9 d for BVDV-2, 200.2 6 116.7 d for BRSV, 65.1 6 37.8 d for replacement product in providing neonatal calves with adequate BHV-1, and 183.8 6 100.0 d for BPIV-3 (14). Another study found the serum levels of antibodies against common respiratory viruses, mean duration of colostrum-derived antibodies in beef calves to be including BVDV-1, BVDV-2, BRSV, BHV-1, and BPIV-3. Two doses 185.6 6 59.8 d for BVDV-1, 157.8 6 56.1 d for BVDV-2, 183.7 6 33 d of the product used in this study provided serum levels of virus- for BRSV, 122.9 6 46.6 d for BHV-1, and 190.6 6 58.3 d for BPIV-3 specific antibodies that were similar to those in calves that received (12). In the present study the mean time to reach seronegativity for 3.8 L of good-quality maternal colostrum (11,13). In addition, the BVDV-1, BVDV-2, BRSV, BHV-1, and BPIV-3 varied with the virus calves that received the replacement product had greater mean but was similar for each virus between the groups. However, the serum titers of antibody against BVDV-1 and BVDV-2 during the first variability in duration was greater in the MC group than in the 4 mo of life compared with the calves that receive maternal colos- CR group: although the standard deviation for time to seronegativ- trum. Higher mean concentrations of BVDV-specific antibodies in ity ranged from 1.5 mo (45 d) to 1.7 mo (53 d) in the MC group, it the colostrum-replacement product and low BVDV-specific antibody ranged from only 0.6 mo (18 d) to 0.8 mo (24 d) in the CR group. concentrations in the maternal colostrum could have contributed to When the proportion of calves becoming seronegative for the various the greater titers of antibody against BVDV-1 and BVDV-2 observed viruses per time period was evaluated, greater proportions of calves in the CR group. The concentrations of BVDV-specific antibodies in the MC group than in the CR group were found to have reached in maternal colostrum could have been influenced by factors such serum titers of antibodies against BVDV-1 and BVDV-2 of # 1:16 as the degree of exposure of cattle within the herd to field strains (log2 # 4), a level associated with an increased susceptibility to of BVDV, the herd vaccination program, the frequency of vaccina- acute BVDV infection (23), at several points during the first months tion, the response to vaccination, and the ability to transfer specific of life. Additionally, at 4 mo of age a greater proportion of calves antibodies to colostrum (11,14,27). The levels of antibodies in the in the MC group than in the CR group had become seronegative to replacement product, in contrast, would be expected to be much BRSV and BHV-1. Variable duration of colostrum-derived immunity

86 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

and differences in the proportion of animals becoming seronega- disease and weight gain in feedlot calves. Can Vet J 1999;40: tive for viral respiratory pathogens could result in poor calf-herd 560–567,570. immunity and increase the risk of introduction of BVDV-1, BVDV-2, 5. O’Connor A, Martin SW, Nagy E, Menzies P, Harland R. The BRSV, and BHV-1 into the calf herd. Variability in the duration of relationship between the occurrence of undifferentiated bovine colostrum-derived immunity against various respiratory viruses has respiratory disease and titer changes to bovine coronavirus and been related to multiple factors, including differences in the rate of bovine viral diarrhea virus in 3 Ontario feedlots. Can J Vet Res decay of colostrum-derived antibodies (11,14,26,27), which is usually 2001;65:137–142. influenced by active viral infections or vaccination. The decay rate of 6. Walz PH, Grooms DL, Passler T, et al. Control of bovine viral passively derived neutralizing antibodies in this study was similar diarrhea virus in ruminants. J Vet Intern Med 2010;24:476–486. in the 2 groups. The calves were not vaccinated at any point, and 7. Moerman A, Straver PJ, de Jong MC, Quak J, Baanvinger T, seroconversion was not observed to any of the viruses, so an active van Oirschot JT. Clinical consequences of a bovine virus diar- viral infection was unlikely. The variation in the levels of colostrum- rhoea virus infection in a dairy herd: A longitudinal study. Vet derived antibodies observed during the first 2 mo of life in the calves Q 1994;16:115–119. in the MC group was likely responsible for the variation in the times 8. Fulton RW, Cook BJ, Step DL, et al. Evaluation of health status to reach seronegative status within this group. of calves and the impact on feedlot performance: Assessment of The results of this study indicate that calves that receive timely a retained ownership program for postweaning calves. Can J Vet and adequate amounts of high-quality maternal colostrum at birth Res 2002;66:173–180. absorb amounts of IgG into the blood that are well above the thresh- 9. Fulton RW, Cook BJ, Blood KS, et al. Immune response to bovine old defined for FPT but still demonstrate highly variable levels of respiratory disease vaccine immunogens in calves at entry to immunity to individual respiratory viruses, including BVDV-1, feedlot and impact on feedlot performance. Bov Practitioner BVDV-2, BRSV, and BHV-1. This results in variable times at which 2011;45:1–12. calves become seronegative for the viral pathogens, which could 10. Ridpath J, Neill J, Endsley J, et al. Effect of passive immunity on increase the risk of acute disease. In contrast to 3.8 L of maternal the development of a protective immune response against bovine colostrum, 2 doses of the colostrum-replacement product provided viral diarrhea virus in calves. Am J Vet Res 2003;64:65–69. calves with adequate passive transfer and less variable levels of 11. Munoz-Zanzi C, Thurmond M, Johnson W, et al. Predicted ages passive immunity to BVDV-1, BVDV-2, BRSV, and BHV-1, which of dairy calves when colostrum derived bovine viral diarrhea may have resulted in the greater uniformity in the time at which virus antibodies would no longer offer protection against disease the calves became seronegative for the viral pathogens. The present or interfere with vaccination. J Vet Med Assoc 2002;221:678–685. study did not attempt to evaluate the ability of passive immunity 12. Kirkpatrick JG, Step DL, Payton ME, et al. Effect of age at the derived from maternal colostrum or a colostrum-replacement time of vaccination on antibody titers and feedlot performance product to protect calves against challenge with virulent strains of in beef calves. J Am Vet Med Assoc 2008;233:136–142. respiratory viruses or to determine the antibody levels associated 13. Fulton RW, Briggs RE, Payton ME, et al. Maternally derived with protection against each virus. Nevertheless, from the results humoral immunity to bovine viral diarrhea virus (BVDV) 1a, it is reasonable to suggest that, overall, calves fed this product will BVDV1b, BVDV2, bovine herpesvirus-1, parainfluenza-3 virus, have longer immunity compared with calves fed the more variable bovine respiratory syncytial virus, Mannheimia haemolytica maternal colostrum, and therefore programs of vaccination against and Pasteurella multocida in beef calves, antibody decline by respiratory viruses should start earlier in calves that receive mater- half-life studies and effect on response to vaccination. Vaccine nal colostrum, as the calves could become seronegative for those 2004;22:643–649. pathogens earlier in life. The more uniform time to seronegativity 14. Kirkpatrick J, Fulton RW, Burge LJ, Dubois WR, Payton M. suggests that it should be possible to better predict the optimum Passively transferred immunity in newborn calves, rate of anti- time to vaccinate calves fed the replacement product used in this body decay, and effect on subsequent vaccination with modified study compared with maternal colostrum, and this should be the live virus vaccine. Bov Practitioner 2001;35:47–54. subject of additional study. 15. Godden SM, Haines DM, Hagman D. Improving passive transfer of immunoglobulins in calves. I: Dose effect of feeding a com- References mercial colostrum replacer. J Dairy Sci 2009;92:1750–1757. 16. Godden SM, Haines DM, Konkol K, Peterson J. Improving 1. Griffin D. Economic impact associated with respiratory disease in passive transfer of immunoglobulins in calves. II: Interaction beef cattle. Vet Clin North Am Food Anim Pract 1992;13:367–377. between feeding method and volume of colostrum fed. J Dairy 2. Stokka GL. Prevention of respiratory disease in cow/calf opera- Sci 2009;92:1758–1764. tions. Vet Clin North Am Food Anim Pract 2010;26:229–241. 17. Pithua P, Godden SM, Wells SJ, Oakes MJ. Efficacy of feeding 3. Ellis JA, Gow SP, Goji N. Response to experimentally induced plasma-derived commercial colostrum replacer for the preven- infection with bovine respiratory syncytial virus following tion of transmission of Mycobacterium avium subsp. paratubercu- intranasal vaccination of seropositive and seronegative calves. losis in Holstein calves. J Am Vet Med Assoc 2009;234:1167–1176. J Am Vet Med Assoc 2010;236:991–999. 18. Weaver DM, Tyler JW, VanMetre DC, Hostetler DE, Barrington 4. Martin SW, Nagy E, Armstrong D, Rosendal S. The associa- GM. Passive transfer of colostral immunoglobulins in calves. tions of viral and mycoplasmal antibody titers with respiratory J Vet Intern Med 2000;14:569–577.

2000;64:0–00 The Canadian Journal of Veterinary Research 87 FOR PERSONAL USE ONLY

19. Quigley JD 3rd, Kost CJ, Wolfe TM. Absorption of protein and 24. Durham PJ, Sillars HM. Evaluation of an enzyme-linked IgG in calves fed a colostrum supplement or replacer. J Dairy immuno­sorbent assay (ELISA) for serodiagnosis of infectious Sci 2002;85:1243–1248. bovine rhinotracheitis infection, with results of a preliminary 20. Chelack BJ, Morley PS, Haines DM. Evaluation of methods for survey. N Z Vet J 1986;34:27–30. dehydration of bovine colostrum for total replacement of normal 25. Durham PJ, Hassard LE. Prevalence of antibodies to infec- colostrum in calves. Can Vet J 1993;34:407–412. tious bovine rhinotracheitis, parainfluenza 3, bovine respira- 21. Walz PH, Givens MD, Cochran A, Navarre CB. Effect of dexa- tory syncytial, and bovine viral diarrhea viruses in cattle in methasone administration on bulls with a localized testicular Saskatchewan and Alberta. Can Vet J 1990;31:815–820. infection with bovine viral diarrhea virus. Can J Vet Res 2008; 26. Van der Poel WHM, Midel WGJ, Schukken YH. Antibody titers 72:56–62. against bovine respiratory syncytial virus in colostrum-fed dairy 22. Passler T, Walz PH, Ditchkoff SS, Walz HL, Givens MD, calves born in various seasons. Am J Vet Res 1999;60:1098–1101. Brock KV. Evaluation of hunter-harvested white-tailed deer 27. Menateau-Horta AM, Ames TR, Johnson DW, Meiske JC. Effect for evidence of bovine viral diarrhea virus infection in Alabama. of maternal antibody upon vaccination with infectious bovine J Vet Diagn Invest 2008;20:79–82. rhinotracheitis and bovine viral diarrhea vaccines. Can Comp 23. Bolin SR, Ridpath JF. Differences in virulence between two non- Med 1985;49:10–14. cytopathic bovine viral diarrhea viruses in calves. Am J Vet Res 1992;53:2157–2163.

88 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY Article

Effect of rider experience and evaluator expertise on subjective grading of lameness in sound and unsound sports horses under saddle Fernando J. Marqués, Cheryl Waldner, Stephen Reed, Fernando Autet, Louise Corbeil, John Campbell

Abstract The primary objective of this study was to investigate whether rider experience influences the assessment and grading of lameness in horses based on under-saddle gait analysis. Thirteen adult sports horses in active training were included in the study. After a baseline lameness and neurologic examination by the principal investigators, horses were videotaped while being ridden by an experienced and a less experienced rider. A 3-minute video was made for each horse and rider and 26 videos were randomly ordered and compiled on a DVD. Veterinarians with different levels of experience in evaluating lameness and veterinary students viewed the DVD and assigned a lameness score to each horse/rider combination. In a model accounting for the expertise of the evaluator, there was no difference in overall lameness scores between experienced and less experienced riders. This result was consistent for both sound and unsound horses. The overall lameness scores reported by specialists and students, however, differed significantly. The lameness score reported by the study participants while the horse was ridden was significantly associated with the subjective baseline lameness assessment reported by the principal investigators for the same limb when the horse was not under saddle. Additional work is necessary to determine whether riders with even lower skill levels would further alter the balance and motion pattern of the horse and have more influence on subjective grading of lameness.

Résumé L’objectif principal de la présente étude était d’examiner si l’expérience du cavalier influence l’évaluation et la gradation de la boiterie chez des chevaux basées sur l’analyse de la démarche sous-selle. Treize chevaux sportifs adultes en entraînement actif ont été inclus dans l’étude. Suite à un examen neurologique et de boiterie de base par les chercheurs principaux, les chevaux ont été filmés alors qu’ils étaient montés par un cavalier d’expérience et un cavalier moins expérimenté. Une vidéo de 3 minutes a été réalisée pour chaque combinaison cheval/cavalier et les 26 vidéos ont été compilées de manière aléatoire sur un DVD. Des vétérinaires ayant différents niveaux d’expérience à évaluer les boiteries et des étudiants vétérinaires ont visualisé le DVD et donné un pointage à chaque combinaison cheval/cavalier. Dans un modèle tenant compte de l’expérience de l’évaluateur, il n’y avait aucune différence dans les pointages globaux entre un cavalier expérimenté et un moins expérimenté. Ce résultat était constant autant pour les chevaux solides que fragiles. Toutefois, les pointages globaux de boiterie notés par les spécialistes et les étudiants étaient significativement différents. Le pointage de boiterie rapporté par les participants à l’étude alors que le cheval était conduit était significativement associé avec l’évaluation subjective de base rapportée par les chercheurs principaux pour le même membre lorsque le cheval n’était pas sous-selle. Des études additionnelles sont nécessaires afin de déterminer si des cavaliers avec encore moins d’expérience influenceraient encore plus l’équilibre et le patron de mouvement du cheval et aurait plus d’influence sur la gradation subjective de la boiterie. (Traduit par Docteur Serge Messier)

Introduction these methods can be impractical in a practice setting and should be used in conjunction with clinical findings (7,8). Visual gait assessment and lameness grading are critical to the Evaluation under saddle is an important component of identifi- diagnosis and ongoing evaluation of musculoskeletal diseases cation, diagnosis, and ongoing assessment in the initial lameness in horses (1–4). Previous studies have shown that the American examination and during treatment (3). Certain orthopedic clinical Association of Equine Practitioners (AAEP) lameness grading system signs and musculoskeletal conditions become apparent only when has a low degree of interobserver agreement (1,2,5,6). There are also the horse is ridden (3,9). In addition, lameness is often evaluated and modern quantitative and more objective techniques for gait assess- monitored for improvement based on performance (3,4). Kinematic ment such as kinematic gait analysis. While there should be less parameters, including gait characteristics, are highly influenced by variability between observers with objective kinematic analysis of the instructions given by the rider (behavioral effects) as well as gait and movement characteristics using more precise measurements, body positioning of the rider (inertial effects) (7,9,10,11). The effect of

Department of Large Animal Clinical Sciences, Western College of Veterinary Medicine, University of Saskatchewan, Saskatoon, Saskatchewan S7N 5B4 (Marqués, Waldner, Corbeil, Campbell); Rood & Riddle, Equine Hospital, Lexington, Kentucky 40580, USA (Reed); and Private Practice, Bs. As., Argentina (Autet). Address all correspondence to Dr. Fernando J. Marqués; telephone: (306) 966-7098; fax: (306) 966-7159; e-mail: [email protected] Received January 11, 2013. Accepted April 23, 2013.

2014;78:89–96 The Canadian Journal of Veterinary Research 89 FOR PERSONAL USE ONLY

the rider on the horse’s motion pattern could interfere with the abil- Out of the 32 horses initially evaluated, only 13 horses met the ity of veterinarians to correctly attribute a disturbed motion pattern inclusion criteria and were included in the study: 6 horses in the to either an intrinsic process in the horse, such as musculoskeletal “lame” group and 7 in the “sound” group. abnormalities, or to external influences, such as a lack of synchrony between horse and rider (9). However, the degree to which the rider Baseline lameness and neurologic examinations influences subjective evaluations of lameness under saddle, either Baseline lameness and neurologic examinations were conducted by disguising or enhancing gait abnormalities, has not been well- by 2 equine practitioners (F.M. and F.A.). Neurologic examination documented. We hypothesized that lameness grading under saddle evaluated the horse’s mentation and behavior, posture, signs of will be altered because of rider experience. muscle mass asymmetry or atrophy, and gait. Horses were then The primary objective of this study was to determine whether assessed for signs of cranial nerve deficits, ataxia, proprioceptive the rider’s experience influences the gait and lameness score of deficits, paresis, spasticity, or dysmetria. sound and unsound horses. The second objective was to determine Lameness examination by the PIs at the study baseline included whether or not gait abnormalities are scored differently by veterinary a distant examination, palpation and manipulation of the limbs specialists, equine practitioners, and veterinary students. The final and back, hoof testing, and upper and lower limb flexion tests. The objective was to compare the results of the traditional baseline lame- horses were examined at a distance to detect any abnormality in ness examination by the principal investigators (PIs) to the lameness conformation, posture, or symmetry. Each horse’s gait was evaluated score reported by the students and participating veterinarians for at a walk and a trot in a 25-m straight line on a flat hard surface, the same limb when the horse was under saddle. while being led with a loose lead shank to allow free motion of the head and neck. Palpation was conducted in the following order: the Materials and methods forelimbs, the neck, the back, and pelvic regions, and then the hind limbs. The limbs were evaluated in a systematic fashion, starting at the hoof and proceeding proximally, including assessing bony Study design structures, soft tissues, and joints for evidence of swelling, localized Thirty-two sports horses in active training were initially screened heat, deformities, effusion, degree of passive motion, and pain. Joints and evaluated for abnormal gait by the principal investigators (F.M. and tendinous structures were palpated with the horses bearing and F.A.). Only sound horses and horses with abnormal gait due to weight on the limb as well as with the limb lifted off the ground. mild lameness [American Association of Equine Practitioners (AAEP) Hoof testers were applied to the frog and the sole from heel to toe lameness score 1 or 2] during the baseline examination were included on both sides of the hoof, assessing elicited pain. Limb flexions in the study. Horses with an AAEP lameness score of $ 3 during the were conducted for 1 min and horses were immediately evaluated baseline examination were excluded from the study due to concerns at a trot in hand in a straight line moving away from the examiner. about purposely riding and making an obviously lame horse perform Any exacerbation of lameness or lameness not previouly seen and a series of potentially painful exercises. Horses were also excluded observed over at least 3 strides was classified as a positive response. from the study if neurologic examination revealed deficits. All horses Finally, the PIs conducted a baseline assessment of the horse’s gait were privately owned, more than 4 y old, and active in dressage, under saddle to detect and grade any gait abnormality, alteration of show jumping, hunter jumping, or western pleasure. Consent was the cranial or caudal phase of the stride, or presence of an abnormal obtained from the owners. The study was approved by the Animal limb flight. The summary baseline gait analysis was recorded based Research Ethics Board at the University of Saskatchewan and fol- on the previously mentioned examination from a distance, hands-on lowed the guidelines of the Canadian Council on Animal Care. examination, and under-saddle gait evaluation. An overall baseline Nine trainers and owners agreed to participate and selected horses lameness score (AAEP lameness score) was assigned for each horse in active training. Horses were included in the “lame” group if at the after all gait and locomotor patterns were assessed. Findings of the time of the baseline examination: 1) they had been diagnosed or had lameness and neurologic examinations were recorded on a stan- been treated for lameness by a veterinarian (other than the authors) dardized form. Baseline data from each limb of each horse used in within the previous 3 mo; and 2) the owner reported intermittent subsequent analyses included whether lameness was observed at a lameness due to this condition within the previous 30 d. In addition, walk and trot in hand, positive or negative flexion test, and positive in order to be included in the “lame” group, the horses had to have or negative hoof tester result. no neurologic deficits, a 1 or 2 out of 5 AAEP lameness score, and a positive response to hoof tester or a positive flexion test that were Rider selection consistent with the previous diagnosis when the principal investiga- All horses in each group were ridden by 1 inexperienced and 1 tors carried out lameness and neurologic examinations. Horses were experienced rider. Riders designated as the lower level experience included in the “sound” group if their owner or trainer reported no group (“inexperienced riders”) had ridden horses for a maximum of history of lameness in the previous 30 d and there were no abnormal 1 y, spent a similar number of hours riding (maximum of 45 min rid- findings during the baseline lameness or neurologic examinations ing time per day and no more than 3 riding days per week), and had for this study. Horses were therefore classified as sound if they had completed a comparable number of formal lessons (maximum of 20). no abnormalities on palpation, flexion tests, or hoof testing, and no Riders that participated as part of the experienced group (“experi- lameness was noted by the PIs during hand walking or trotting or enced riders”) were also selected to be as similar as possible in terms when evaluated under saddle (AAEP lameness score 0). of riding ability and number of years riding (minimum of 5 y) and

90 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

After neurologic and lameness examinations are performed, the following riding pattern is to be carried out. 1. a. Walk from point A to point B and back. b. Travel in a posting trot from point A to point B and back.

Point A 30 m (98 ft) Point B 2. a. At a walk, travel around entire perimeter of arena on the Right rein (clockwise) until back at starting location. Then repeat on Left rein (counterclockwise). b. Same pattern as above but in a posting trot. c. Same pattern as above but in a sitting trot. d. Same pattern as above but in an extended posting trot. 40 m (131 ft)

20 m (66 ft)

3. a. Complete one 10-m diameter circle at a sitting trot, traveling clockwise (Right rein). b. Complete one 10-m diameter circle at a sitting trot, traveling counterclockwise (Left rein).

10-m 10-m diameter diameter

4. a. At a sitting trot, travel in a loose figure-8 pattern using the whole arena and crossing the middle of the arena on the long diagonal. b. Same pattern as above but at an extended trot. c. Same pattern as above but at a canter, with 2 simple lead changes in the center of the arena when traveling across the diagonal (i.e., bring the horse back to a trot).

5. a. Travel in a counterclockwise circle at the sitting trot using an entire half of the arena. After one circle has been completed, switch reins while in the center of the arena and continue on the opposite rein in a circle using the other half of the arena (i.e., a figure-8 pattern). b. Same pattern as above but at a canter, coming back to a trot in the center of the arena before changing to the opposite rein.

Figure 1. Riding patterns.

2000;64:0–00 The Canadian Journal of Veterinary Research 91 FOR PERSONAL USE ONLY

ness analysis [4 American College of Veterinary Surgeons (ACVS) Diplomates, 3 ACVS senior surgery residents, and 5 equine practi- tioners], as well as 5 third-year veterinary students with a strong sports-horse background. They were blinded to the selection criteria and baseline examination results for the horses and were asked to identify any gait abnormality and assign a lameness score to indi- vidual limbs at each specific gait and an overall lameness score based on all gaits and locomotor patterns. Since the AAEP scoring system grades lameness at both the walk and the trot combined and is therefore not suitable for assessing indi- vidual limbs at specific gaits (walk only, trot only and canter only), a 3-grade lameness score was used for that purpose. The 3-grade lameness score was used to grade the individual limbs at each spe- cific gait (walk, sitting trot, rising trot, and canter). This lameness score was defined as: score 0 — lameness not perceptible (sound); grade 1 — lameness difficult to observe; and grade 2 — lameness Figure 2. Still picture from one of the video footages. The videos were consistently observable. edited such that the rider’s body was blanked out. The rider’s aids and The study participants used a standard 5-grade AAEP lameness commands given to the horse were not seen and therefore the riding skills of the riders were not obvious to the study participants. scoring system to assign an overall lameness score for each limb combining all gaits. Grades 0, 1, and 2 of the 3-grade lameness score system used to assess individual limbs at single specific gaits cor- had succesfully competed (placed 1st, 2nd, or 3rd at a sports horse responded to the same grades (0, 1, and 2) of the traditional AAEP show) at a high level within their sport during the previous 5 y. The lameness score system used to assess individual limbs combining selected riders were also of similar body weights. Five experienced all gaits. A standard form was used to record the video evaluation riders, 2 of whom were professional riders, and 4 inexperienced rid- lameness scores for each horse and rider combination recorded in ers met the inclusion criteria and agreed to participate in the study. the footage. Videotaping horses under saddle Statistical analysis Immediately after the baseline lameness and neurologic examina- All data were entered, recoded, and summarized in a commer- tions were completed, the 13 horses that met the inclusion criteria for cial spreadsheet program (Microsoft Excel; Microsoft, Redmond, the “lame” or “sound” groups were videotaped while being ridden Washington, USA). The lameness scores measured in this study by 1 experienced and 1 inexperienced rider through a standardized should be considered as ordinal data. Ordinal logistic regression series of locomotor patterns. The locomotor patterns and riding exer- models, however, would not converge after considering the impor- cises were designed to enable the horses to be evaluated at a walk, tant fixed and random effects. trot, and canter as well as to incorporate lateral flexion (Figure 1). Therefore to address the first 2 objectives, linear mixed models Both the experienced and inexperienced riders were selected from (PROC MIXED; SAS, Version 9.2, SAS Institute, Cary, North Carolina, the riding facility where the horses were stabled. The order in which USA) were used to evaluate the association of the following variables the experienced and inexperienced riders rode the horses was sys- with overall lameness score for each limb (n = 1768 observations) as tematically randomized. Each horse was ridden by the 2 riders with reported by the study participants after viewing the video footage only the minimum time necessary to allow 1 rider to dismount and for each horse and rider combination. The variables examined as the second rider to mount the horse. The riders were blinded to the fixed effects in the model included horse soundness as assessed by history and baseline lameness examination results for all horses. the PIs, rider experience, the interaction between rider experience All horses were videotaped in the ring at their riding facility from and soundness, and evaluator experience. Limb position, reported the same angle and at the same distance from the video camera. as front compared to hind limbs, was also evaluated as a potential Eleven horses were ridden in 1 of 4 different rings with sand footing modifier for the effect of rider experience. The model accounted and 2 horses were ridden in 1 of 2 different rings with grass footing. for the lack of independence in the data associated with multiple The arenas were all similar in size, with a minimum of 20 3 40 m measurements using random intercepts for limb (n = 4 per horse- (approximately 65.6 3 131 ft). rider combination in each video footage), unique horse and rider combination captured in the video footage (n = 13 horses 3 2 riders), Video editing and lameness evaluation by study and evaluator (n = 17). The results of the models were reported as participants the difference associated with category of rider experience and then The videos were edited to exclude the rider from the frame the difference associated with category of observer experience, as (Figure 2). A 3-minute video was made for each horse and rider well as the least squared mean lameness scores with 95% confidence combination and all footages (total 26) were randomly ordered intervals (CIs) for each category (95% CI). and compiled on a DVD. Each footage was assessed by a group of Variance partition coefficients (VPCs) were estimated to determine veterinarians with varying expertise in sports medicine and lame- the contributions of the random effects to the final model using the

92 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

Table I. Baseline lameness examination results obtained by the principal investigators (PI). A lameness score was assigned to each horse while hand-walking and hand-trotting in a 25-m straight line, in addition to an overall lameness score following the AAEP guidelines. The “lame” group was comprised of horses with a lameness history, an overall AAEP score of 1 or 2, as well as either a positive upper or lower limb flexion test or a positive hoof test response (horses ID # 6, 10, 14, 18, 19, and 25) Overall AAEP Lameness score: Lame lameness score Horse walk and trot in a straight line Flexion tests Hoof test under and limb ID LF RF LH RH LF RF LH RH LF RF saddle assigned by PI 2 0 0 0 0 2 2 2 2 2 2 no 0 6 0 0 0 0 2 2 2 2 1 1 yes 1 LH 9 0 0 0 0 2 2 2 1 2 2 no 0 10 0 0 0 0 2 2 1 1 2 1 yes 1 LH 11 0 0 0 0 2 2 2 2 2 2 no 0 12 0 0 0 0 2 2 2 2 2 2 no 0 13 0 0 0 0 2 2 2 2 2 2 no 0 14 0 0 2 0 2 2 1 2 2 2 yes 2 LH 18 0 0 0 0 2 2 1 1 2 2 yes 1 LH 19 2 0 0 0 1 1 2 2 2 2 yes 2 LF 21 0 0 0 0 2 2 2 2 2 2 no 0 25 0 0 2 0 2 2 1 1 2 2 yes 2 LH 26 0 0 0 0 2 2 2 1 2 2 no 0 LF — left front limb; RF — right front limb; LH — left hind limb; RH — right hind limb; AAEP — American Association of Equine Practitioners.

2 2 2 following formulas: 1) VPClimb = s limb/(s limb 1 s horse*rider combination 1 of the generalized linear mixed models were reported as odds ratios 2 2 2 s evaluator); 2) VPChorse*rider combination = s horse*rider combination/(s limb 1 (OR) with 95% CI. 2 2 2 s horse*rider combination 1 s evaluator); and 3) VPCevaluator = s evaluator/ For all models with continuous outcomes, the residuals and ran- 2 2 2 (s limb 1 s horse*rider combination 1 s evaluator). dom effects (best linear unbiased predictors) were graphed against significant predictors to assess the homogeneity of the variance. To evaluate the third objective, linear mixed models for continuous The residuals and random effects at each level were also examined outcomes (PROC MIXED, SAS, Version 9.2, SAS Institute) were again graphically to determine whether the residuals were adequately used to evaluate whether baseline assessments by the PIs (numeric normally distributed. Finally, for both continous and dichotomous value) were associated with the overall lameness score for each limb outcomes, the residuals and random effects were graphed against reported by study participants based on assessment of the video the predicted values to determine whether there were any extreme footages (n = 1768 observations). The following baseline results from outliers or influential results. the principal investigors were considered as fixed effects in a series of models for overall limb lameness score: 1) the numeric value of lameness score; 2) whether or not the PIs reported a lameness score Results of 2 at a walk and trot in hand (1 or 0); 3) positive flexion test for the limb (1 or 0); and 4) positive hoof tester for the limb (1 or 0). The limb position (left fore, left hind, right fore, right hind) and exper- Selected horses tise of the assessor were also considered as potential confounders Horses from various breeds and activities were included in the in all models. This model also included random intercepts for limb study. All horses were in active training and stabled at private rid- (n = 4 per horse-rider combination in each video footage), unique ing facilities around Saskatoon, Saskatchewan, Canada and Buenos horse and rider combination captured in the video footage (n = 13 Aires, Argentina. Five horses (38%) were Warmbloods, 3 (23%) horses 3 2 riders), and evaluator (n = 17). were American Paints, 2 (15%) were Quarter horses, 2 (15%) were In the second step for the third objective, generalized linear mixed Thoroughbreds, and 1 (8%) was Arabian. Of the 13 horses partici- models (PROC GLIMMIX, Version 9.2; SAS Institute) with a logit pating in the study, 6 (46%) were competing in show jumping at link function and binomial distribution for dichotomous outcomes an intermediate level, 6 (46%) were used for English and Western (whether or not the participant score was $ 2) were evaluated using pleasure riding, and 1 (8%) was a top performance endurance horse. the same fixed and random effects. The results of the mixed models All horses were trained at the same facility where they were stabled, were reported as regression coefficients with 95% CI and the results examined, and videotaped. Nine horses (69%) were female and

2000;64:0–00 The Canadian Journal of Veterinary Research 93 FOR PERSONAL USE ONLY

4 (31%) were geldings. Horses varied in age from 4 to 18 y [mean: 7.4; study participants while the horse was being ridden, after accounting standard deviation (SD): 4.0; median: 7]. for the limb evaluated and the experience of the evaluator (n = 1768). The baseline lameness score for the limb reported by the PIs was Summary of lameness examination results associated with the participant lameness score based on the video The PIs assigned 6 of the 13 horses with an overall AAEP baseline footage. The mean participant lameness score increased by 0.58 units lameness score of 1 or 2, in addition to $ 1 positive result on flexion (95% CI: 0.39 to 0.78, P = 0.0001) for every 1 unit increase reported by testing or hoof testing (Table I). At a trot, 1 out of 13 horses was lame the PIs. Similarly, whether or not the PIs assigned the limb a lame- in the left front limb (LF) and 2 were lame in the left hind limb (LH). ness score of 2 was also associated with the participants’ score for One of 13 horses tested positive in a flexion test of the LF, 1 tested the limb. The mean lameness score assigned by study participants positive in a flexion test of the right front limb (RF), 4 tested positive increased by 1.17 score points (95% CI: 0.78 to 1.56, P = 0.0001) when in a flexion test of the LH, and 5 tested positive in a flexion test in the PI baseline lameness score was 2 compared to when it was $ 2. the right hind limb (RH). When assessed with hoof testers, 1 horse Finally, if the PIs considered the limb to be positive on flexion test, tested positive in the LF and 2 horses tested positive in the RF. the mean lameness score of study participants increased by 0.33 score Based on the overall AAEP scores from the evaluation of video points (95% CI: 0.08 to 0.57, P = 0.01). The results of a hoof tester footage of each horse, the results among the study participants examination were not associated with the lameness scores for the were much more variable. Maximum scores assigned to any limb limb reported by the participants (P = 0.88). among all study participants ranged from 0 to 2 out of 5 for 1 horse, We also considered whether or not the baseline lameness assess- 0 to 3 for 6 horses, 0 to 4 for 3 horses, 1 to 3 for 2 horses, and 2 to 4 ment by the PIs was associated with an increased likelihood that for 2 horses. the study participants would report a lameness score of $ 2 for the same limb when viewing the video footage of the horse under saddle Rider experience and reported lameness score (n = 1768). Limbs classified by the PIs with a baseline lameness score from video footage of 2 (OR: 14.5, 95% CI: 8.2 to 25.4, P = 0.0001) were more likely to be In a model (n = 1768) accounting for the expertise of the evalua- assigned a lameness score of $ 2 by study participants observing the tor, the effect of rider experience on the overall lameness score for animals being ridden after accounting for limb and observer experi- each limb did not vary whether or not the horse was considered ence. Similarly, limbs reported as positive on the flexion test were sound, i.e., there was no interaction between rider experience and also more likely to be given a score of $ 2 by study participants (OR: the horse’s classification as sound or lame on entry to the study 4.0, 95% CI: 2.6 to 6.1, P = 0.0001). There was no significant associa- (P = 0.72). There was also no interaction between rider experience tion, however, between the hoof tester results and the likelihood that and whether the lameness was observed in the front or hind limb study participants would describe the limb as having a score of $ 2 (P = 0.40). (OR: 3.1, 95% CI: 0.8 to 12.4, P = 0.11). In a model accounting only for the expertise of the evaluator, there was no difference in overall lameness scores between experienced Discussion and inexperienced riders (difference 0.017, 95% CI: 20.065 to 0.097, P = 0.70). The mean lameness score was 0.42 (95% CI: 0.28 to 0.55) Although other studies have demonstrated the importance of for the “inexperienced rider group” and 0.40 (95% CI: 0.26 to 0.54) the interaction between the rider and the horse and its effect on the for the “experienced rider group.” horse’s gait and motion (9,10,12–14), the objective of this study was to determine to what extent riders with different levels of experience Evaluator experience and reported lameness can influence subjective evaluations of under-saddle lameness, by scores from video footage either disguising or enhancing gait abnormalities. Rider experience There were no significant differences in the overall lameness was not associated with lameness scores in either sound or unsound scores reported by practitioners and specialists (difference 20.05, horses in this analysis. This differs from previous reports that sug- 95% CI: 20.24 to 0.14, P = 0.60) or those reported by practitioners and gested subtle differences in the balance of the rider’s weight could students (difference 0.15, 95% CI: 20.05 to 0.35, P = 0.14) (n = 1768). mask or exacerbate lameness and that poor riding can potentially However, there was a significant difference in the overall lameness lead to a false-positive diagnosis of lameness (3). The absence of a scores reported by specialists and students (difference 0.20, 95% CI: rider experience effect in this study could have been the result of low 0.01 to 0.39, P = 0.04). The mean lameness scores were 0.44 (95% CI: study power due to a limited number of horses (lame and sound) 0.26 to 0.62) for the “practitioners group,” 0.49 (95% CI: 0.33 to 0.65) and rider combinations or it could be a real finding. The relatively for the “specialists group,” and 0.29 (95% CI: 0.11 to 0.47) for the tight confidence intervals for the difference between experienced “students group.” The VPC was 0.03 for limb, 0.002 for horse and and less experienced riders, however, suggest that the power of the rider combination, and 0.22 for evaluator. In the final model, the study was adequate for the range of rider experience and the extent proportion of variance not explained by the random effects was 0.74. and types of lameness included in this study. The finding in this study that reported lameness did not differ Association between PI baseline evaluation and with rider experience could also be due to smaller than expected participant evaluation of video footage variation in the ability of our experienced and less experienced Most measures of baseline lameness reported by the PIs were riders. Without riding errors such as substantial unbalanced dis- associated with the lameness score for the same limb reported by the tribution of weight or inappropriate shift of center of gravity, our

94 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

inexperienced riders would not have altered the horse’s balance and this type of research. Although in our study we accounted for many motion pattern in such a way as to influence the lameness grading. of these factors, such as the expertise of the evaluator, effect of rider Less skilled riders could be less balanced, which could alter the gait experience, limb evaluated, and whether the horse was classified of the horses enough to influence the lameness grading. While the as lame or sound, many other influences on the horse’s gait and inexperienced riders had only very basic riding skills, in general the motion pattern could not be measured and considered in the analysis. PIs subjectively perceived them to be off balance for only very short Investigating a larger population of lame horses with a wider range periods of time, while overall their weight distribution and dynamic of lameness severity and location within and among limbs is war- balance were adequate. ranted. It would also be valuable to evaluate the effect of horse use The expertise of the professionals evaluating lameness is also and the type of lameness and to explore differences among individual important to consider when using subjective scores to identify subtle riders and the type of arena footing more specifically. Furthermore, changes in the horse’s balance and motion pattern. There were no assessing a larger number of horses would also make it possible to significant differences in the overall lameness scores reported by determine whether the rider’s ability influences subjective grading of practitioners and specialists or those reported by practitioners and lameness depending on the diagonal on which the rider sits in rising students. The overall lameness scores reported by specialists and those trot and the presence of lameness in the front versus the hind limb. reported by students, however, differed significantly. The specialist In conclusion, rider experience did not change subjective grading group consisted of 4 ACVS Diplomates with many years of experience of lameness in the present study when videotaped footage of horses in evaluating lame horses and 3 ACVS residents with less experience. under saddle was assessed. In a model accounting for the expertise The practitioners group included equine veterinarians from the equine of the evaluator, however, there was a significant difference in the field service unit, 1 with more than 10 y and the rest with 2 to 4 y overall lameness scores reported by ACVS Diplomates and residents of experience in equine practice. The similarity in overall lameness and scores from senior veterinary students. Additional research scores reported by the specialist and practitioner groups could reflect studies are necessary in order to determine whether riders with even the fact that the ACVS senior residents that were included in the lower skill levels would further alter the balance and motion pattern specialists group and the veterinarians with 2 to 4 y experience from of the horse and have a more apparent influence on the subjective the equine field service unit that were included in the practitioners grading of lameness. group had comparable lameness identification skills. Furthermore, the students group included third year veterinary students with a Acknowledgments strong sports-horse and riding background who could be expected to have observation skills approaching some of the less experienced This study was funded by the Western College of Veterinary practitioners. Only the differences in the overall lameness scores Medicine (WCVM) Research Trust Fund at the University of reported by specialists and students were statistically significant. It Saskatchewan. The authors thank Peter Downing and Wayne is also possible that lameness scores between evaluators would have Giesbrecht of Media Access and Production (eMAP) at the University differed more if horses with a broader range of lameness severity, such of Saskatchewan for assistance in editing and compiling the video as AAEP grade 3, had been included in the study. The very small vari- and Diego J. Roiz of Haras Villanueva for his collaboration with the ance partition coefficients associated with limb and horse and rider equitation technical support. combinations, however, suggest that differences between horses did not account for a substantial part of the unexplained variation in the References final model for observer lameness score. Previous studies have shown that there is a low degree of interob- 1. Guide to Veterinary Services for Horse Shows. 7th ed. Lexington: server agreement on subjective gait analysis and lameness grading American Association of Equine Practitioners, 1999. (2,5,6). In our study, the subjective lameness assessments by the PIs 2. Stashak TS. Examination for lameness. In: Stashak TS, ed. were significantly associated with the lameness score perceived by Adams’ Lameness in Horses. 5th ed. Philadelphia, Pennsylvania: the study participants for each limb while the horse was being rid- Lippincott Williams and Wilkins, 2002:113–183. den. Limbs classified by the PIs as having a lameness score of 2 while 3. Ross MW. Movement. In: Ross MW and Dyson SJ, eds. Diagnosis the horses were not being ridden were more likely to be assigned a and Management of Lameness in the Horse. St. Louis, Missouri: lameness score of $ 2 by study participants observing the horses being WB Saunders, 2003:72–79. ridden. Similarly, limbs reported as positive on the flexion test were 4. Keegan KG. Evidence-based lameness detection and quantifica- also more likely to be given a score of 2 or more by study participants. tion. Vet Clin North Am Equine Pract 2007;23:403–423. Although there was no association between the hoof tester results and 5. Keegan KG, Wilson DA, Wilson DJ, et al. Evaluation of mild the likelihood that participants would describe the limb as having a lameness in horses trotting on a treadmill by clinicians and score of $ 2 in the present study, the very small number of positive interns or residents and correlation of their assessments with hoof tester results severely limited our ability to address this question. kinematic gait analysis. Am J Vet Res 1998;59:1370–1377. The intrinsic nature of subjectivity when assessing lameness with 6. Hewston M, Christley RM, Hunt ID, Voute LC. Investigations of the naked eye, as well as multiple factors affecting the horse’s gait the reliability of observational gait analysis for the assessment and motion, i.e., rider’s skills, balance, and weight, behavioral effects of lameness in horses. Vet Rec 2006;158:852–857. of the rider, instructions given by the rider, as well as type, cause, 7. Ishihara A, Reed SM, Rajala-Schultz PJ, Robertson JT, Bertone location, and severity of lameness, makes it challenging to conduct AL. Use of kinetic gait analysis for detection, quantification, and

2000;64:0–00 The Canadian Journal of Veterinary Research 95 FOR PERSONAL USE ONLY

differentiation of hind limb lameness and spinal ataxia in horses. with experimentally induced forelimb lameness. Am J Vet Res J Am Vet Med Assoc 2009;234:644–651. 2005;66:1805–1815. 8. Weishaupt MA, Wiestner T, Hogg HP, Jordan P, Auer JA, 12. Symes D, Ellis R. A preliminary study into rider asymmetry Barrey E. Assessment of gait irregularities in the horse: Eye within equitation. Vet J 2009;181:34–37. vs. gait analysis. Eq Vet J Suppl 2001;33:135–140. 13. Blokhuis M, Aronsson A, Hartmann E, Van Reenen C, Keeling L. 9. Peham C, Licka T, Schobesberger H, Meschan E. Influence of the Assessing the rider’s seat and horse’s behavior: Difficulties and rider on the variability of the equine gait. Hum Mov Sci 2004; perspectives. J App Anim Welf Sci 2008;11:191–203. 23:663–671. 14. Schöllhorn W, Peham C, Licka T, Scheidl M. A pattern recogni- 10. Powers P, Harrison A. Effects of the rider on the linear kinematics tion approach for the quantification of horse and rider interac- of jumping horses. Sports Biomech 2002;1:135–146. tions. Equine Vet J Suppl 2006;36:400–405. 11. Ishihara A, Bertone AL, Rajala-Schultz PJ. Association between subjective lameness grade and kinetic gait parameters in horses

96 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY Article

Acute exercise does not induce an acute phase response (APR) in Standardbred trotters Lena Kristensen, Rikke Buhl, Katarina Nostell, Lars Bak, Ellen Petersen, Maria Lindholm, Stine Jacobsen

Abstract The purpose of the study was to investigate whether acute strenuous exercise (1600- to 2500-m race) would elicit an acute phase response (APR) in Standardbred trotters. Blood levels of several inflammatory markers [serum amyloid A (SAA), haptoglobin, fibrinogen, white blood cell count (WBC), and iron], muscle enzymes [creatinine kinase (CK) and aspartate transaminase (AST)], and hemoglobin were assessed in 58 Standardbred trotters before and after racing. Hemoglobin levels increased and iron levels decreased 12 to 14 h after racing and haptoglobin concentrations, white blood cell counts, and iron levels were decreased 2 and/or 7 d after racing. Concentrations of CK, AST, SAA, and fibrinogen were unaltered in response to racing. Acute strenuous exercise did not elicit an acute phase reaction. The observed acute increase in hemoglobin levels and decreases in haptoglobin and iron levels may have been caused by exercise-induced hemolysis, which indicates that horses might experience a condition similar to athlete’s anemia in humans. The pathogenesis and clinical implications of the hematological and blood- biochemical changes elicited by acute exercise in Standardbred trotters in the present study warrant further investigation.

Résumé L’objectif de la présente étude est d’investiguer l’effet de l’exercice intensif (1600–2500 m) sur l’apparition d’une phase inflammatoire aigué chez des trotteurs. Les taux sanguins de différents marqueurs inflammatoires [serum amyloid A (SAA), haptoglobine, fibrinogène, globules blanc (WBC) et fer], d’enzymes musculaires [creatinine kinase (CK) et aspartate transaminase (AST)] et d’hémoglobine ont été évalué dans 58 trotteurs avant et après la course. Respectivement une augmentation et diminution des taux d’hémoglobine et de fer furent présente 12–14 heures suivant la course. Une réduction des taux d’haptoglobine, de globules blanc et de fer était présente 2 et/ou 7 jours après la course. Aucun changement dans les concentrations de CK, AST, SAA et fibrinogène n’a pu être démontre en relation avec la course. L’exercice intense réalise lors de cette étude n’a pu induire de phase inflammatoire aigué. L’augmentation d’hémoglobine et la diminution d’haptoglobine et de fer peuvent être causées par une hémolyse induite par l’exercice, indiquant qu’une condition similaire à l’anémie de l’athlète chez l’homme existe chez le cheval. La pathogénèse et l’implication clinique des modifications hématologiques et biochimiques lié à un exercice intense chez les trotteurs dans la présente étude justifié de plus amples investigations. (Traduit par Docteur Denis Verwilghen)

Introduction change during the inflammatory response. While WBC will increase during inflammation, serum iron concentrations will decrease (3,4). The acute phase response (APR) is the rapid and non-specific It has been shown that strenuous exercise elicits a short-lived APR response elicited by all sorts of inflammatory stimuli, such as infec- in humans and horses, with increases in WBC and serum concentra- tion, trauma, cancer, and ischemia. It is characterized by release of tions of APPs. Most of these studies have investigated the APR in signal molecules from injured cells and tissues; these molecules trig- endurance sports such as 120- to 160-km endurance races in horses ger synthesis of pro-inflammatory cytokines that will induce hepatic (5,6) and marathon, ultramarathon, and triathlon competitions in synthesis of acute phase proteins (APPs). The APPs are released into humans (7). There are, however, studies in humans, dogs, and horses the blood stream, where they can be measured in increased concen- that have investigated the effect of acute exercise of a shorter dura- trations (1). In the horse, several proteins have acute phase proper- tion at high intensity. Two studies of trained human amateur runners ties, e.g., serum amyloid A (SAA), fibrinogen, and haptoglobin. In (8,9) investigated the response to shorter running distances (5 and response to inflammatory stimuli, blood concentrations of fibrinogen 15 to 21 km), as did a study in sled dogs (10) performing moderate- and haptoglobin will increase modestly (up to 10 times healthy duration, high-intensity exercise (racing 16 miles each of 2 days). values), while SAA concentrations may increase several 100- or These studies showed that shorter running distances also caused 1000-fold (2). Along with levels of APPs, levels of other parameters an APR with small and short-lived increases in WBC and serum such as white blood cell counts (WBC) and iron concentrations will concentrations of C-reactive protein, the major human and canine

Holstebro Equine Practice, Vinderup, Denmark (Kristensen); Department of Large Animal Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Taastrup, Denmark (Buhl, Bak, Petersen, Lindholm, Jacobsen); Department of Clinical Sciences, Faculty of Veterinary Medicine and Animal Science, Swedish University of Agricultural Sciences, Uppsala, Sweden (Nostell). Address all correspondence to Dr. Stine Jacobsen; telephone: 145 35332620; fax: 145 35332880; e-mail: [email protected] Received August 31, 2012. Accepted April 9, 2013.

2014;78:97–102 The Canadian Journal of Veterinary Research 97 FOR PERSONAL USE ONLY

APP. In contrast, a recent study of horses participating in limited Table I. Levels of creatinine kinase (CK), aspartate distance endurance rides (34 or 60 km) showed no post-race change transaminase, acute phase proteins, iron, hemoglobin, and in APPs (6). The APR after acute, short-duration exercise (sprinting) white blood cell count in 26 Standardbred trotters before has been investigated to a limited degree in horses. One study of (day -1) and after (12 to 14 h) racing 1640 or 2140 m. Levels Thoroughbreds showed no SAA response 1 h after a 1600-m race are shown as mean 6 standard deviation (range). Within each (11). No studies are available on the effects of racing on the APR of row, different superscript letters signify statistical difference Standardbred trotters. (P , 0.05) between sampling times The APR to exercise seems to be related to running distance (9) Day -1 14 h and degree of muscle injury, as measured by serum concentrations Creatinine kinase (U/L) 316.1 6 368.3a 262.2 6 79.84a of creatinine kinase (CK) (12,13). Training has been shown to cause a (157 to 2035*) (166 to 454) reduction in the post-run APR in humans (14), thus suggesting that an improved physical condition protects against the exercise-induced Aspartate transaminase (U/L) 434.8 6 188.4a 465.6 6 156.5a APR. In contrast, a previous study in Thoroughbreds demonstrated (172.0 to 1021) (170.0 to 825.0) increased fibrinogen and haptoglobin concentrations during the Serum amyloid A (mg/L) 1.6 6 0.23a 1.6 6 0.36a last weeks of an 80-d training program. It was suggested that this (1.1 to 2.0) (1.2 to 2.8) could be an effect of both the adjustments to training and subclinical Fibrinogen (g/L) 3.87 6 1.47a 3.54 6 1.13a disease becoming manifest over time (15). (1.98 to 5.73) (1.89 to 6.01) The hypothesis of this study was therefore that acute, strenuous exercise (1600 to 2500 m-race) induces an APR in Standardbred Iron (mmol/L) 29.09 6 8.93a 25.47 6 6.17b trotters. Assessing levels of acute phase reactants could potentially (15.48 to 59.47) (15.44 to 37.36) yield information about individual fitness and the possible need for Hemoglobin (g/L) 138.7 6 10.12a 150.6 6 11.77b a prolonged rest after racing. (118.0 to 157.0) (130.0 to 170.0) White blood cell count 7.27 6 1.24a 7.86 6 1.31a Materials and methods (3 109/L) (5.6 to 9.9) (5.6 to 10.0) * One horse had elevated CK before racing, but in a control sample Blood samples were obtained from 58 Standardbred trotters obtained 1 h after racing, levels were 600 U/L (data not shown) and [22 stallions, 13 mares, 23 geldings; mean age 4.8 y (from 3 to 11 y)]. the measured value may be an error or result from contamination of Pre-race samples (day -1) were collected from all horses. Samples the phlebotomy needle with tissue. As no reason for excluding this were collected 12 to 14 h after racing (group A, n = 26) or 2 and 7 d outlier was identified, it was retained in the analyses to avoid bias. after racing (group B, n = 32). The horses were all fit racehorses in professional training that raced at different racetracks in Denmark and Sweden. Racing distances were 1600 m (n = 1), 1640 m (n = 11), reduction of nicotinamide-adenine dinucleotide phosphate (NADP) 1700 m (n = 5), 1800 m (n = 2), 1900 m (n = 7), 2000 (n = 1), 2020 m to nicotinamide-­adenine dinucleotide phosphate plus hydrogen (n = 1), 2100 m (n = 5), 2140 m (n = 18), and 2500 m (n = 6), all of (NADPH), and by AST-catalyzed L-aspartate and 2-oxoglutarate which are considered sprinting distances. Racing distance was not reaction with subsequent maleate dehydrogenate-catalyzed NADH recorded for 1 horse. Information about racing speed (kilometer time, oxidation indicator reaction] according to International Federation i.e., time in seconds when trotting a 1-km distance) and finishing of Clinical Chemistry guidelines in an ADVIA 1800 Chemistry position was recorded for each horse. System (Siemens Health Care Diagnostics, Deerfield, Illinois, On the day of blood collection, each horse underwent a general USA). Serum SAA concentrations were determined by a previously clinical examination. Blood was collected in tubes (Vacutainer validated immunoturbidometric method (LZ test SAA; EIKEN Systems; Becton Dickinson, Meylan, France) containing sodium- Chemical, Tokyo, Japan) (16). Haptoglobin was assessed by colori- ethylene diamine triacetic acid (EDTA) for determining WBC and metric spectrophotometry (Phase Range Haptoglobin Assay; Tridelta hemoglobin, tubes containing sodium citrate for determining plasma Development, Maynooth, Ireland) on an automated ADVIA 1800 fibrinogen, and tubes with no additive for preparing serum samples Chemistry System (Siemens Health Care Diagnostics) as described for analyzing concentrations of SAA, iron, CK (only measured in in a previous study (17). Fibrinogen concentrations were determined group A), aspartate transaminase (AST) (only measured in group A), by the Clauss method in an automated coagulometric analyzer and haptoglobin (only measured in group B). Citrated plasma was (ACL 9000; Instrumentation Laboratory, Barcelona, Spain). Serum prepared by centrifugation at 2000 3 g for 10 min. Serum was iron concentrations were determined by colorimetric spectropho- prepared by letting blood samples coagulate for approximately 6 h tometry (ADVIA 1650; Bayer A/S, Lyngby, Denmark). before centrifugation at 2000 3 g for 10 min. Analyses were carried Different analysis techniques were used for the 2 groups of horses out immediately (WBC, hemoglobin) or serum/plasma was stored to determine WBC and hemoglobin concentrations. In group A, for up to 4 wk at 220°C before analyses were carried out (CK, AST, hematological analyses were done using an ADVIA 2120 analyzer SAA, haptoglobin, fibrinogen, and iron). (Siemens Healthcare Diagnostics) with equine settings. In group B, Serum CK and AST were determined spectrophotometrically 2 handheld analyzers (HemoCue Hb 2011 and HemoCue WBC; [by CK-catalyzed N-acetylcysteine-activated reaction with adenos- HemoCue, Vedbæk, Denmark) were used according to the manu- ine diphosphate (ADP) and, through several steps, subsequent facturer’s instructions.

98 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

Table II. Levels of acute phase proteins, iron, hemoglobin, and white blood cell count in 32 Standardbred trotters before (day -1) and after (day 2 and 7) racing 1600 to 2500 m. Levels are shown as mean 6 standard deviation (range). Within each row, different superscript letters signify statistical difference (P , 0.05) between sampling times Day -1 Day 2 Day 7 Serum amyloid A (mg/L) 10.4 6 42.2a 0.8 6 4.2a Below DLa (below DL-214*) (below DL-25.1) (all below DL) Haptoglobin (g/L) 1.91 6 0.25a 1.11 6 0.29b 1.15 6 0.24a,b (0.70 to 1.59) (0.52 to 1.55) (0.73 to 1.59) Fibrinogen (g/L) 3.1 6 0.4a 3.1 6 0.4a 3.0 6 0.4a (2.2 to 4.0) (2.3 to 3.9) (2.3 to 4.2) Iron (mmol/L) 31.98 6 7.8a 31.88 6 8.2a 28.35 6 9.2b (18.81 to 52.13) (16.98 to 54.38) (16.29 to 60.60) Hemoglobin (g/L) 149.4 6 19.17a 150.6 6 23.77a 148.4 6 15.59a (79.0 to 193.0) (96.0 to 214.0) (98.0 to 175.0) White blood cell count (3 109/L) 8.8 6 2.2a 8.3 6 1.4a,b 7.7 6 1.5b (5.2 to 14.5) (5.8 to 10.7) (5.7 to 11.2) DL — detection limit of the assay (corresponding to 0.5 mg/L SAA). * It was not clear why 1 horse had mildly elevated SAA before racing, as the horse was clinically healthy and SAA levels were low in the subsequent sample. As no reason for excluding this outlier was identified, it was retained in the analyses to avoid bias.

All statistical analyses were carried out using Prism 4.02 Racing distance and racing speed [kilometer time; average and (GraphPad, San Diego, California, USA). Normal distribution of data standard deviation (SD) 76.2 6 2.2, range 72.0 to 84.5 s/km] were was assessed by D’Agostino and Pearson omnibus normality test and generally not related to magnitude of change in levels of the blood all parameters, except SAA, were found to be normally distributed. parameters, but in group B, WBC levels decreased to a greater Changes over time in WBC and serum levels of CK, AST, SAA, hap- magnitude in horses racing 2500 m than in horses racing shorter toglobin, fibrinogen, and iron within each group were evaluated by distances (P , 0.05). paired t-test (group A) or repeated measurements analysis of variance (ANOVA) with pairwise comparisons by Bonferroni test for signifi- Discussion cant effects (group B). Due to the different laboratory analyses used for determining some of the measured parameters in the 2 groups of Acute, high-intensity exercise caused changes in the levels of iron, horses, levels could not be compared between the 2 groups. The cor- hemoglobin, haptoglobin, and WBC in the blood of the horses. The relation between blood parameters and kilometer time was evaluated results of the present study did not, however, support the hypoth- by linear regression. Differences in blood parameters between racing esis that exercise induces an inflammatory response. Levels of the distances were evaluated by Student’s t-test (group A) or one-way 4 positive acute phase parameters remained unchanged (SAA and ANOVA and pairwise comparisons (Bonferroni) for significant effects fibrinogen) or decreased (haptoglobin, WBC) after racing. Serum (group B). A 5% significance level was used. amyloid A concentrations will increase several 100-fold even after mild inflammatory stimuli such as minor arthroscopic procedures Results (18), foaling (19), or vaccination (20). Even minor injuries to the horses during racing, such as muscle damage, that cause a sub- All horses remained clinically healthy throughout the study, sequent APR would thus likely have resulted in increased serum except for 1 horse that was excluded from the study as it developed SAA concentrations. The horses in the present study did not seem a hoof abscess. Five horses galloped during the race and their kilo- to suffer muscle damage, as serum levels of CK and AST were not meter times were thus not included in further analyses. significantly elevated after racing. This is in contrast to horses per- Analyses showed that, in group A, iron decreased significantly forming endurance rides of 120 or 160 km, which had elevated serum (P , 0.05) and hemoglobin increased significantly (P , 0.05), while CK and SAA levels after the race (5,6). Two studies on the effects CK, AST, fibrinogen, SAA, and WBC remained unchanged (Table I). of training on CK and APPs in Thoroughbreds and Standardbreds In group B, haptoglobin, iron, and WBC decreased significantly showed increases in CK and fibrinogen and haptoglobin over the (P , 0.01, P , 0.05, and P , 0.01, respectively) (Table II), while SAA, course of an 80-d training ­program (15,21). These differences are fibrinogen, and hemoglobin concentrations remained unchanged very likely related to differences in duration of exercise and type of throughout the study period (Table II). exertion between studies.

2000;64:0–00 The Canadian Journal of Veterinary Research 99 FOR PERSONAL USE ONLY

Timing of sampling is important in order not to miss a potential The increase in blood hemoglobin levels demonstrated in group A increase in SAA and other APPs. Previous studies have shown 12 to 14 h after racing and the decrease in serum haptoglobin con- that serum concentrations of SAA will increase within 8 to 24 h centration demonstrated in group B 2 days after racing may indicate of an inflammatory stimulus, while fibrinogen and haptoglobin that subclinical hemolysis occurred in response to the exercise, which responses develop more slowly (18,20,22). No significant change in is similar to findings in numerous studies in humans (28,30–34). SAA levels were detected in Standardbreds 1 h after completing a Haptoglobin associates strongly with hemoglobin released from standardized 1600-m race (11), but due to the kinetics of the SAA red blood cells during hemolysis. Decreased serum haptoglobin response, increases in SAA concentrations would not be expected concentrations have thus been demonstrated in horses after intra- with this short interval between stimulus and sampling. In contrast, vascular hemolysis (35), and 2 studies have shown that hemoglobin the sampling intervals in the current study (12 to 14 h, 2 and 7 d levels increase and haptoglobin levels decrease in Thoroughbreds after the race) were long enough to allow SAA, fibrinogen, and after individual training sessions or over the course of a 3-wk train- haptoglobin responses to develop, if present (2). Therefore, the ing program (26,36). As hemoglobin in the present study was total lack of APP response cannot be ascribed to the choice of sampling hemoglobin, not plasma hemoglobin, increases in numbers of red times. blood cells, e.g., in response to splenic contraction, may also have The WBC decreased 7 d after racing. The pathogenesis of this affected the results. decrease is not clear. Taking into account the kinetics of leukocyte Hemolysis and changes in iron metabolism such as those demon- response (when fast decreases in leukocyte numbers occur only in strated in the present study are thought to be the cause of athlete’s response to severe and peracute inflammation such as sepsis), it is anemia. This condition, which is characterized by persistently low not likely that the small, yet statistically significant decrease on day 7 levels of blood hemoglobin, low hematocrit, and low ferritin con- is the result of race-related inflammation. Leukocyte numbers, popu- centrations, is well-described in humans, particularly in athletes lations, and functions may change in response to acute exercise (23). practicing endurance sports. It has been suggested that this condition Increases in leukocyte numbers are usually observed within hours of causes suboptimal performance or even fatigue. Several factors may acute exercise and the significance and interpretation of the late and contribute to its development in humans, including hemolysis caused small decrease in WBC observed in the present study are thus not by the foot striking against hard ground, gastrointestinal bleeding, clear (23,24). The consequences, if any, of such changes in terms of hematuria, sweating, and menstruation in females (37). Additional performance and health of the horses are currently unknown, but in studies are needed with more detailed analyses of factors involved humans, heavy training has been linked to immunosuppression and in erythrocyte turnover and iron metabolism to determine whether increased incidence of upper airway infections (23). Further studies anemia occurs in equine athletes. It is not clear if and how changes are needed to determine whether changes in leukocyte numbers in erythrocyte-related parameters, haptoglobin levels, and iron status are a consistent feature of acute strenuous exercise in horses and to affect the equine athlete clinically or in terms of performance. investigate whether the observed small changes have any clinical In conclusion, the acute exercise performed in the present study significance. did not elicit any kind of inflammatory response with changes in The measurements of the 4 positive acute phase parameters thus levels of acute phase parameters. This is in contrast to studies in did not support the hypothesis of a racing-induced inflammatory endurance horses after 120- to 160-km rides (5,6), and is most likely response. In contrast, the decrease in serum iron levels observed in explained by the short racing distance and different type of exercise both groups of horses might reflect an exercise-induced inflamma- performed by the Standardbreds (1600- to 2500-m racing). For race- tory response. Iron is a very sensitive marker of inflammation with a horses, races of this distance are considered sprinting and the effect fast kinetic. It will decrease within hours after an inflammatory insult of sprinting on the APR has not previously been investigated in and return to baseline levels 1 to 2 d after inflammation has waned Standardbred trotters. In contrast, blood-biochemical changes were (18,20). Inflammatory cytokines such as interleukin-1b, ­interleukin-6, demonstrated that were consistent with acute hemolysis (increased and tumor necrosis factor a are released and cause hepatic upregula- hemoglobin and decreased haptoglobin and iron), which might tion of the hormone hepcidin during inflammation (4) as well as dur- suggest that even short bouts of strenuous exercise by the horses ing exercise (25). This has a negative influence on iron transportation can lead to changes similar to those observed in human athletes and and absorption and, as such, exercise-induced inflammation might horses after acute short or longer term exercise (6,30,36). It needs to potentially lead to a drop in serum iron concentrations similar to that be determined whether these changes affect the performance and/or demonstrated in other inflammatory conditions in horses (20). In the health of the horses. absence of other inflammatory signs, however, non-inflammatory causes of decreased serum iron concentrations are more likely in Acknowledgments the present study. Hemolysis and/or volume expansion have been suggested as potential causes of decreased serum iron in horses The authors thank all trainers and horse owners for participating undergoing a 3-wk training program (26). In humans, many exercise- in this study. The study was funded by grants from the Anniversary related factors can affect serum iron concentrations, e.g., loss of iron Foundation of the Kingdom of Denmark Horse Insurance, The Kustos in sweat (27), increased erythrocyte destruction with subsequent Foundation of 1881, Boehringer Ingelheim Denmark, Merial Norden, hemoglobinuria (28), and gastrointestinal bleeding (29). It was not and E-vet Denmark. HemoCue (Vedbæk, Denmark) kindly provided determined whether one or more of these mechanisms contributed the test strips for determining WBC and hemoglobin concentrations to iron loss in the horses involved in the present study. in group B horses.

100 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

References 18. Jacobsen S, Nielsen JV, Kjelgaard-Hansen M, et al. Acute phase response to surgery of varying intensity in horses: A preliminary 1. Gabay C, Kushner I. Acute-phase proteins and other systemic study. Vet Surg 2009;38:762–769. response to inflammation. New Engl J Med 1999;340:448–454. 19. Nunokawa Y, Fujinaga T, Taira T, et al. Evaluation of serum 2. Jacobsen S, Andersen PH. The acute phase protein serum amy- amyloid A protein as an acute-phase reactive protein in horses. loid A (SAA) as a marker of inflammation in horses. Eq Vet Edu J Vet Med Sci 1993;55:1011–1016. 2007;19:38–46. 20. Andersen S, Petersen HH, Ersbøll AK, Falk-Rønne J, Jacobsen S. 3. Peeling P. Exercise as a mediator of hepcidin activity in athletes. Vaccination elicts a prominent acute phase response in horses. Eur J Appl Physiol 2010;110:877–883. Vet J 2012;191:199–202. 4. Oliveira JP, Badial PR, Cunha PHJ, et al. Lipopolysaccharide- 21. Fazio F, Assenza A, Tosto F, Casella S, Piccione G, Caola G. infusion up-regulates hepcidin mRNA expression in equine liver. Training and haematochemical profile in Thoroughbreds Innate Immun 2012;18:436–446. and Standardbreds: A longitudinal study. Livest Sci 2011;141: 5. Cywinska A, Gorecka R, Szareska E, Witkowski L, Dziekan P, 221–226. Schollenberger A. Serum amyloid A level as a potential indicator 22. Jacobsen S, Niewold ThA, Halling-Thomsen M, et al. Serum of the status of endurance horses. Eq Vet J 2010;42:23–27. amyloid A isoforms in serum and synovial fluid in horses with 6. Cywinska A, Szarska E, Gorecka R, et al. Acute phase protein lipopolysaccharide-­induced arthritis. Vet Immunol Immunopathol concentrations after limited distance and long distance endur- 2005;110:325–330. ance rides in horses. Res Vet Sci 2012;93:1402–1406. 23. Walsh NP, Gleeson M, Shepard RJ, et al. Position statement. 7. Kasapis C, Thompson PD. The effects of physical activity on Part one: Immune function and exercise. Exerc Immunol Rev serum C-reactive protein and inflammatory markers. A system- 2011;17:6–63. atic review. J Am Coll Cardiol 2005;45:1563–1569. 24. Lejeune JP, Sandersen C, Votion D, et al. Effect of intensive 8. Drenth JPH, Krebbers RJM, Bijzet J, van der Meer JWM. exercise on plasmatic neutrophil elastase level in eventing and Increased circulating cytokine receptors and ex vivo ­interleukin-1 endurance horses. Eq Vet J 2010;42:12–16. receptor antagonist and interleukin 1b production but decreased 25. Peeling P, Dawson B, Goodman C, Landers G, Trindr D. Athletic tumour necrosis factor-a production after a 5-km run. Eur J Clin induced iron deficiency: New insights into the role of inflam- Invest 1998;28:866–872. mation, cytokines and hormones. Eur J Appl Physiol 2008;103: 9. Strachan AF, Noakes TD, Kotzenberg G, Nel AE, de Beer FC. 381–391. C reactive protein concentration during long distance running. 26. Inoue Y, Matsui A, Asai Y, Aoki F, Matsui T, Yano H. Effect of Br Med J 1984;289:1249–1251. exercise on iron metabolism in horses. Biol Trace Elem Res 10. Wakshlag JJ, Kraus MS, Gelzer AR, Downey RL, Vacchani P. The 2005;107:33–42. influence of high-intensity moderate duration exercise on cardiac 27. Paulev P-E, Jordal R, Pedersen NS. Dermal excretion of iron in troponin I and C-reactive protein in sled dogs. J Vet Intern Med intensely training athletes. Clin Chim Acta 1983;127:19–27. 2010;24:1388–1392. 28. Dufaux B, Hoederath A, Streitberger I, Hollmann W, Assman G. 11. Giori L, Moretti P, Giordano A, Paltrinieri S. Short-term evalu- Serum ferritin, transferrin, haptoglobin, and iron in middle- ation of serum amyloid A after exercise in clinically healthy and long-distance runners, elite rowers, and professinal racing horses. J Eq Vet Sci 2011;31:499–501. cyclists. Int J Sports Med 1981;2:43–46. 12. Taylor C, Rogers G, Goodman C, et al. Hematologic, iron-related, 29. Buckman MT. Gastrointestinal bleeding in long-distance runners. and acute phase protein responses to sustained strenous exercise. Ann Int Med 1984;101:127–128. J Appl Physiol 1987;62:464–469. 30. Martinez AC, Escanero JF. Iron, transferrin, and haptoglobin 13. Weight LM, Alexander D, Jacobs P. Strenous exercise: Analogous levels after a single bout of exercise in men. Physiol Behav 1992; to the acute-phase response? Clin Sci (London) 1991;81:677–683. 51:719–722. 14. Liesen H, Dufaux B, Hollmann W. Modifications of serum 31. Peeling P, Dawson B, Goodman C, et al. Training surface and glycoproteins the days following a prolonged exercise and the intensity: Inflammation, hemolysis, and hepcidin expression. influence of physical training. Eur J Appl Physiol Occup Physiol Med Sci Sports Exerc 2009;41:1138–1145. 1977;37:243–254. 32. Peeling P, Dawson B, Goodman C, et al. Cumulative effects of 15. Fazio F, Assenza A, Tosto F, Casella S, Piccione G, Caola G. consecutive running sessions on hemolysis, inflammation and Modifications of some acute phase proteins and the white blood cell hepcidin activity. Eur J Appl Physiol 2009;106:51–59. count in thoroughbreds during training. Vet Rec 2010;167:370–372. 33. Telford RD, Sly GJ, Hahn AG, Cunningham RB, Bryant C, 16. Jacobsen S, Kjelgaard-Hansen M, Petersen HH, Jensen AL. Smith JA. Footstrike is the major cause of hemolysis during Evaluation of a commercially available human serum amyloid A running. J Appl Physiol 2003;94:38–42. (SAA) turbidometric immunoassay for determination of equine 34. Yusof A, Leithauser RM, Roth HJ, Finkernagel H, Wilson MT, SAA concentrations. Vet J 2006;172:315–319. Beneke R. Exercise-induced hemolysis is caused by protein 17. Pihl TH, Andersen PH, Kjelgaard-Hansen M, Mørck NB, modification and most evident during the early phase during Jacobsen S. Serum amyloid A and haptoglobin in serum and an ultraendurance race. J Appl Physiol 2007;102:582–586. peritoneal fluid of healthy horses and horses with acute abdomi- 35. Allen B, Archer RK. Haptoglobins in the horse. Vet Rec 1971;89: nal pain. Vet Clin Pathol 2013;42:177–183. 106–109.

2000;64:0–00 The Canadian Journal of Veterinary Research 101 FOR PERSONAL USE ONLY

36. Cywinska A, Szarska E, Kowalska A, Ostaszewski P, 37. Chatard J-C, Mujika I, Guy C, Lacour J-R. Anemia and iron Schollenberger A. Gender differences in exercise-induced intra- deficiency in athletes. Practical recommendations for treatment. vascular haemolysis during race training in Thoroughbred Sports Med 1999;27:229–240. horses. Res Vet Sci 2011;90:133–137.

102 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY Article

Recovery from desflurane anesthesia in horses with and without post-anesthetic xylazine Turi K. Aarnes, Richard M. Bednarski, Alicia L. Bertone, John A.E. Hubbell, Phillip Lerche

Abstract The objective of this study was to compare recovery from desflurane anesthesia in horses with or without post-anesthetic xylazine. Six adult horses were anesthetized on 2 occasions, 14 d apart using a prospective, randomized crossover design. Horses were sedated with xylazine, induced to lateral recumbency with ketamine and diazepam, and anesthesia was maintained with desflurane. One of 2 treatments was administered intravenously at the end of anesthesia: xylazine [0.2 mg/kg body weight (BW)] or an equivalent volume of saline. Recovery parameters were recorded and assessed by 2 blinded observers. A Wilcoxon signed-rank test was used to analyze recovery data. Heart rate, arterial blood pressures, and arterial blood gas data were analyzed using 2-way analysis of variance (ANOVA) for repeated measures. Values of P , 0.05 were considered significant. Duration of anesthesia was not different between groups. Administration of xylazine at the end of desflurane anesthesia was associated with significantly longer times to first movement, endotracheal tube removal, first attempt to achieve sternal recumbency, sternal recumbency, first attempt to stand, and standing. Number of attempts to stand and quality of recovery scores were not different between groups. Administering xylazine after desflurane anesthesia resulted in longer recovery times. Recovery scores were not significantly different between groups.

Résumé L’objectif de la présente étude était de comparer la récupération suite à une anesthésie au desflurane chez des chevaux avec ou sans administration post-anesthésie de xylazine. Six chevaux adultes furent anesthésiés à deux occasions à 14 j d’intervalle, en utilisant un design expérimental croisé aléatoire. Les chevaux ont été soumis à une sédation à la xylazine, mis en décubitus latéral avec de la kétamine et du diazépam, et l’anesthésie maintenue avec du desflurane. Un des deux traitements suivants fut administré par voie intraveineuse à la fin de l’anesthésie : xylazine (0,2 mg/kg de poids corporel) ou un volume équivalent de saline. Les paramètres de récupération furent enregistrés et évalués à l’aveugle par deux observateurs. Le test de comparaison des données de Wilcoxon fut utilisé pour analyser les données de récupération. Le rythme cardiaque, la pression artérielle, et les données des gaz sanguins artériels furent analysés par analyse de variance (ANOVA) pour des mesures répétées. Des valeurs de P , 0,05 étaient considérées comme significatives. La durée de l’anesthésie n’était pas différente entre les groupes. L’administration de xylazine à la fin de l’anesthésie au desflurane était associée à des délais significativement plus longs avant : un premier mouvement, le retrait du tube endotrachéal, un premier essai pour se mettre en décubitus sternal, le décubitus sternal, un premier essai pour se mettre debout, et se tenir debout. Le nombre d’essais pour se tenir debout et la qualité des pointages de récupération n’étaient pas différents entre les groupes. L’administration de xylazine suite à l’anesthésie au desflurane a entraîné des temps de récupération plus longs. Les pointages de récupération n’étaient pas significativement différents entre les groupes. (Traduit par Docteur Serge Messier)

Introduction rapidly eliminated by the lungs on cessation of administration (6,7). Desflurane is associated with a more rapid return to consciousness Anesthesia is associated with a greater risk of increased morbid- than other available inhalational anesthetics because it has the lowest ity and mortality in horses than in other frequently anesthetized blood-gas solubility, but shortening recovery time may not result in domestic species (1–6). Since problems during recovery from anethe- improved recovery characteristics (8). The administration of seda- sia, such as fractures and myopathies, account for approximately tives, including xylazine, to horses before or during recovery from 30% of the morbidity and mortality associated with inhalational inhalational anesthesia has been recommended in order to postpone anesthesia in the horse, it is important to carefully manage the initial attempts to stand and to improve recovery (9–11). return to standing (6). Inhalational anesthetic drugs are preferred The use of desflurane in the horse has been described, but the for procedures lasting more than 60 min in the horse because they drug has not been widely adopted because of expense and the are minimally metabolized, do not accumulate over time, and are requirement for a complex, electrically heated vaporizer for safe

Department of Veterinary Clinical Sciences, The Ohio State University College of Veterinary Medicine, Columbus, Ohio 43210, USA. Address all correspondence to Dr. Turi Aarnes; telephone: (614) 292-3551; fax: (614) 292-0895; e-mail: [email protected] This paper was presented at the 15th International Veterinary Emergency and Critical Care Symposium and Annual Conference of the American College of Veterinary Anesthesiologists, Phoenix, Arizona, 2008. Received November 13, 2012. Accepted May 6, 2013.

2014;78:103–109 The Canadian Journal of Veterinary Research 103 FOR PERSONAL USE ONLY

administration (12). Recovery characteristics for desflurane are endotracheal tube (internal diameter, 26 mm) and positioned in typical of horses recovering from inhalational anesthesia (13). Post- dorsal recumbency on a foam mattress. The endotracheal tube was anesthetic administration of a combination of xylazine and propo- then connected to a large animal circle system (Mallard Medical, fol after 4 h of desflurane anesthesia in healthy horses resulted in Redding, California, USA) that had been primed by occluding the improved transition from lateral recumbency to standing, but also Y piece and setting the oxygen flow rate at 8 L/min and the des- resulted in respiratory depression (14). The effects of xylazine for flurane vaporizer (Tec 6 Vaporizer; Omeda, Madison, Wisconsin, post-anesthetic sedation have been studied extensively in horses USA) at 8% for 10 min before anesthetic induction, based on mini- anesthetized with isoflurane or sevoflurane (10,11,15,16), but have mum alveolar concentration of desflurane in the horse (13). Horses not been investigated after desflurane anesthesia. In 1 study, admin- breathed spontaneously for 15 min after connection to the circle istration of xylazine to horses after isoflurane anesthesia resulted in system. Horses were mechanically ventilated (6 breaths/min and a lower degree of sedation than in horses that received romifidine approximate tidal volume of 10 to 15 mL/kg BW as estimated from (11). The quality of recovery in isoflurane-anesthetized horses that the bellows housing) beginning 15 min after induction and ventila- received xylazine or romifidine was not different (10,11). The pur- tion was adjusted to maintain the partial pressure of arterial carbon pose of the study reported here was to further compare the recovery dioxide (PaCO2) at between 40 and 50 mmHg. During maintenance, characteristics from desflurane anesthesia in horses with and without the oxygen flow rate was decreased to between 5 and 10 mL/kg BW post-anesthetic sedation with xylazine. Our hypothesis was that per minute. Anesthetic depth was monitored based on changes in post-anesthetic sedation with xylazine would prolong recovery times blood pressure and palpebral and corneal reflexes and the vaporizer and improve recovery scores in horses anesthetized with desflurane. was adjusted to maintain a surgical plane of anesthesia sufficient to prevent movement. Materials and methods A facial artery was catheterized with a 20-gauge, 1 1/4-in catheter to measure arterial blood pressure and to facilitate the collection of heparinized arterial blood for blood gas analysis. Arterial blood Animals pressures were measured using a fluid-filled catheter and pressure The study population consisted of 6 clinically healthy horses transducer (Edwards Life Science, Irvine, California, USA) zeroed (2 Quarter horses and 4 Thoroughbreds; 4 mares and 2 geldings) at the level of the scapulohumeral joint. Dobutamine (Bedford anesthetized as part of a separate study on bone healing. Mean age Laboratories, Bedford, Ohio, USA) was administered as necessary to was 12 y (range: 3 to 24 y) and mean weight was 545 kg (range: maintain mean arterial blood pressure at between 60 and 90 mmHg. 432 to 655 kg). This study was approved by the Ohio State University Lactated Ringer’s solution (Baxter Healthcare) was administered Animal Care and Use Committee. IV at a rate of 5 mL/kg BW per hour until just before horses were moved to the recovery stall. Heart rate, arterial blood pressures, and Study design respiratory measurements were monitored and recorded. Arterial Food but not water was withheld for 12 h before the experiment. blood samples were collected and analyzed (IRMA; Diametrics Each horse was anesthetized on 2 occasions, 14 d apart. During the Medical, St. Paul, Minnesota, USA and Radiometer ABL735; first anesthetic event, osteotomy of a 4th metacarpal or metatarsal Radiometer America, Westlake, Ohio, USA) for determining pH bone was carried out on each limb and 3 holes were created in the and blood gas at 15 min (immediately before mechanical ventila- 12th rib on both sides of the thorax of each horse for a study on the tion), 45 min after anesthetic induction, and every hour thereafter, use of fibroblast-mediated bone morphogenetic protein-2 therapy or at the end of anesthesia. End-tidal desflurane concentration (Poet to accelerate bone healing (17). During the second anesthetic event, IQ2 Model 8500Q; Criticare Systems, Waukesha, Wisconsin, USA) bone growth-stimulating cells were injected into each site created was monitored and recorded for 9 of the 12 anesthetic events, due during the first anesthetic event. All surgeries were done by the same to equipment servicing. The instrument was calibrated before use experienced surgeon (AB), at the same time of day (morning). Each according to manufacturer specifications. A nasopharyngeal tube horse received phenylbutazone (Phenylbutazone; Vedco, St. Joseph, (internal diameter, 12 or 14 mm) was inserted during anesthetic Missouri, USA), 2 g IV, 30 min before surgery and on the first post- maintenance and remained in place until the horse was standing. operative day. After recovery assessment and complete recovery from anesthesia, post-operative pain was treated intramuscularly Recovery with morphine (Morphine; Baxter Healthcare, Deerfield, Illinois, At the end of surgery, bandages were applied around the can- USA), 0.1 mg/kg body weight (BW). non bones of all 4 limbs, the arterial catheter was removed, and the A 14-gauge polytetrafluoroethylene catheter (Intracath; Parke horse was disconnected from the anesthetic machine. At the time of Davis, Sandy, Utah, USA) was inserted in a jugular vein for the disconnection from the anesthetic machine, 1 of 2 treatments was administration of anesthetic drugs and fluids. Horses were sedated randomly administered IV in a blinded crossover design: 0.2 mg/kg with xylazine (X-Ject E; Butler Animal Health Supply, St. Joseph, BW of xylazine or an equivalent volume of 0.9% saline, so that the Missouri, USA), 1 mg/kg BW, IV and anesthesia was induced 3 horses that received xylazine in the first anesthetic event received 5 min later with ketamine (Ketaset; Fort Dodge Laboratories, Fort saline in the second anesthetic event (18). The horse was transported Dodge, Iowa, USA), 2.2 mg/kg BW, IV and diazepam (Diazepam; 10 ft and hoisted from the table to a 4 3 4 m square padded recovery Hospira, Lake Forest, Illinois, USA), 0.1 mg/kg BW, IV combined in stall and positioned in left lateral recumbency. Ambient light and a single syringe. Horses were orotracheally intubated with a cuffed noise were minimized. One person returned to the recovery stall to

104 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

Table I. Median recovery times and quality scores Parameter Control Xylazine Duration of anesthesia (min) 165 (110 to 200) 143 (117 to 208) Duration of surgery (min) 98 (70 to 110) 90 (60 to 95) Time to first movement (min) 17 (13 to 29) 28 (15 to 55)a Time to extubation (min) 14 (13 to 29) 27 (15 to 55)a Time to first attempt to sternal recumbency (min) 23 (19 to 35) 50 (24 to 78)a Time to sternal recumbency (min) 26 (21 to 35) 50 (24 to 78)a Time to first attempt to stand (min) 31 (21 to 56) 51 (29 to 97)a Time to standing (min) 31 (22 to 72) 51 (29 to 97)a Number of attempts to stand 2 (1 to 4) 1 (1 to 2) Recovery score 1.0 (1 to 2.5) 1.0 (0 to 3.0) Data are presented as median (range). a Significant difference (P , 0.05) from control group. remove the endotracheal tube when the horse swallowed and then as median (range). A 1-tailed paired nonparametric t-test (Wilcoxon left the stall. Due to the nature of the bone-healing study and the signed-rank test) was used to analyze the single-variable recovery potential for post-surgical fracture, head and tail ropes were attached data. Heart rate, arterial blood pressures, arterial blood gas, vapor- to the nosepiece of the halter and tied to the tail, respectively as izer setting, and oxygen flow rate data are reported as mean 6 a precautionary measure. The ropes were inserted through rings standard deviation (SD) and were analyzed using 2-way analysis on opposite walls of the recovery stall and experienced personnel of variance (ANOVA) for repeated measures for main effects and allowed slack in the ropes during recovery. Personnel remained interaction. A Bonferroni post test was done to identify differences outside the recovery stall. If the horse moved, slack was taken up to within and between the 2 groups. Values of P , 0.05 were consid- prevent tangling. Tension was applied to the ropes only if person- ered significant. nel felt that injury might occur to the horse based on its behavior during recovery. Results Duration of anesthesia was defined as the time from when the horse was connected to the anesthetic circle to when it was discon- Median recovery times and quality scores are presented in Table I. nected. Time to first movement, time to removal of the endotracheal Anesthetic duration was 143 min for horses that received post- tube, time to first attempt to sternal recumbency, time to sternal anesthetic xylazine sedation and 165 min in the control group, with recumbency, time to first attempt to stand, and time to standing no differences between groups. In addition, time for the first surgi- were recorded from the time the horse was disconnected from the cal procedure was 90 min and 93 min for the second procedure and anesthetic circle. The number of attempts to stand was recorded. was not different between procedures. Compared to saline, post- Recovery was recorded with a video camera and the quality of anesthetic sedation with xylazine was associated with longer times recovery was graded based on a modified version of a quantitative to first movement (28 and 17 min, respectively), extubation (27 and recovery scoring system (7,19,20), which grades recovery on a scale 14 min, respectively), first attempt to achieve sternal recumbency of 0 to 4. Grade 0 corresponded to excellent recovery where the (50 and 23 min, respectively), achieving sternal recumbency (50 and horse stood on the first attempt with clean effort and little/no body 26 min, respectively), first attempt to stand (51 and 31 min, respec- sway/shifting; grade 1 corresponded to good recovery where the tively), and standing (51 and 31 min, respectively). The number horse stood on the first or second attempt with slight body sway/ of attempts to stand was not different between groups. The mean shifting once standing; grade 2 corresponded to fair recovery where recovery scores were not different between groups, and the kappa the horse stood after 2 or 3 attempts with a strong effort on the last statistic was 0.66, indicating substantial inter-evaluator agreement attempt and with body shifting once standing; grade 3 corresponded (21,22). More horses in the sedated group (4 horses) than in the to poor recovery marked by several attempts to stand, marked unsedated group (3 horses) stood on their first attempt and more instability once standing, and possible minor injury to the horse; and horses in the sedated group had a recovery score of 0 (2 horses) than grade 4 corresponded to unacceptable recovery marked by several in the unsedated group (0 horses). No horses or personnel were weak attempts to stand, falling easily or resumed recumbency, and injured and no myopathies or lacerations were noted for any of the possible major injury to the horse. Two independent observers were horses during recovery. While ropes were applied as a precautionary blinded to the treatment and assessed recovery in real time (PL) and measure due to the surgical procedure, tension was not applied to by videotape review (RB). the ropes at any time to assist recovery. There were no differences in cardiovascular, respiratory, blood Analysis of data gas parameters, desflurane vaporizer settings, or oxygen flow rates The kappa statistic was used to evaluate inter-observer agree- between groups (Table II). End-tidal desflurane concentrations ment of recovery scores. Recovery scores were averaged between were monitored and recorded for 9 of the 12 anesthetic events due the 2 investigators. Non-parametric recovery results are reported to equipment servicing and were not statistically evaluated. Each

2000;64:0–00 The Canadian Journal of Veterinary Research 105 FOR PERSONAL USE ONLY 129 (4) 2.1 (4) 140 (4) 1.4 (4) 0.7 (2) 1.7 (3) 0.2 (2) 0 (3) 0 (1) 1 (2) 0 (1) 12 (3) 7 (2) 17 (3) 11 (2) 9 (3) 9 (2) 0.04 (4) 0.06 (4) 11 (3) 6 41 6 39 6 80 6 53 6 67 6 62 6 79 6 8.5 1.9 6 1.8 6 148 6 135 6 100 6 40.4 6 42.2 6 10.0 6 10.5 6 12.8 6 165 min 7.46 6 7.44 6

178 2.5 105 2.9 0.8 1.7 0.2 1.1 8 0.5 (5) 0.7 (4) 15 8 8 12 5 9 0.03 6 0.04 47 6 47 6 98 6 68 6 65 6 83 6 76 6 1.9 6 2.3 6 9.6 6 220 6 213 6 105 6 41.2 6 42.6 6 11.3 6 11.0 6 10.0 6 105 min 7.45 6 7.45 6 — partial pressure of arterial oxygen; — partial pressure of arterial oxygen; 2

1.3 0.9 0.2 1.1 9 0.5 (5) 0.6 (4) 13 17 12 13 11 11 12

6 — — — — — — 46 6 48 6 95 6 71 6 63 6 88 6 74 6 1.9 6 2.3 6 9.2 6 9.6 6 107 6 10.8 11.0 6 90 min

1.9 1.5 0.2 1.1 11 0.9 (5) 0.7 (4) 11 13 6 12 6 13 6

— — — — — — 44 6 46 6 99 6 65 6 64 6 79 6 76 6 1.9 6 2.3 6 8.9 6 9.3 6 103 6 10.3 6 10.5 6 75 min a

2.1 0.2 1.5 8 1.3 1.1 (5) 1.0 (4) 8 8 9 6 5 8 5

— — — — — — 40 6 41 6 97 6 99 6 58 6 59 6 72 6 73 6 2.0 6 2.4 6 8.9 6 9.0 6 10.3 6 10.7 6 60 min

170 1.7 2.4 136 1.1 4 1.4 0.2 1.2 1.1 (5) 0.9 (4) 6 7 7 4 7 5 6 0.03 0.04

37 6 38 6 91 6 93 6 51 6 54 6 66 6 67 6 2.0 6 2.7 6 8.6 6 8.8 6 236 6 228 6 44.1 6 10.8 6 42.6 6 10.0 6 45 min 7.43 6 7.45 6

1.2 3 0.3 1.0 2.4 1.0 (5) 1.0 (4) 8 9 12 7 7 8 9

— — — — — — 37 6 39 6 92 6 89 6 53 6 50 6 66 6 63 6 2.2 6 3.7 6 8.2 6 8.3 6 30 min 10.5 6 10.7 6

1.0 92 6.5 3 57 7.7 1.3 0.5 0.4 1.1 (5) 1.4 (4) 5 8 19 5 12 6 15 0.04 0.04

39 6 37 6 63 6 65 6 78 6 81 6 9.8 6 2.4 6 2.3 6 8.4 6 7.9 6 160 6 125 6 103 6 107 6 10.2 6 65.2 6 67.4 6 7.28 6 7.28 6 15 min

standard deviation. 6 7 6

6 6 43 — — — Awake — — — — — — — 38 — — — — — — — —

Xylazine Xylazine Control Control Group Xylazine Xylazine Control Xylazine Control Xylazine Control Control Xylazine Control Xylazine Control Xylazine Control Control Xylazine

— vaporizer setting; EtDES end-tidal desflurane concentration. — partial pressure of arterial carbon dioxide; Vap

2 2

2

flow O Six horses were used, except where number of is in parentheses (for time point 165 min and EtDES concentrations). 2 aCO ap

P V HR (mmHg) (mmHg) PaCO Table II. Cardiovascular, respiratory, and oxygen flow rates, blood gas, and desflurane variables and oxygen respiratory, II. Cardiovascular, Table Parameter a (%) Data for EtDES and time point 165 minutes are presented, but not analyzed due to insufficient numbers. O (L/min) EtDES (%) (beats/min) Data are presented as mean HR — heart rate; SABP systolic arterial blood pressure; DABP diastolic MABP mean PaO SABP (mmHg) DABP (mmHg) MABP (mmHg) Pa

pH

106 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

horse required dobutamine during each anesthetic event. The dose horses compared with isoflurane-anesthetized horses. It is possible of dobutamine ranged from 0.0625 to 1 mcg/kg BW per min and was that desflurane was eliminated so rapidly at the end of anesthe- varied to maintain a mean arterial blood pressure of 60 to 90 mmHg. sia that the additional time afforded by xylazine to allow further decreases in inhalant concentration was of no apparent benefit, with Discussion horses transitioning successfully from lateral recumbency to standing regardless of whether or not they were sedated. Horses that received In this study, post-anesthetic sedation with xylazine after des- xylazine tended to have better recovery scores in our study. Although flurane anesthesia delayed times to first movement, extubation, this was not statistically different, it is also possible that our results first attempt to achieve sternal recumbency, achievement of sternal might have been different if a larger number of horses had been recumbency, first attempt to stand, and standing. studied. Further evaluation of recoveries of desflurane-anesthetized One-third of the mortalities associated with equine anesthesia are horses that receive post-anesthetic alpha-2 agonists is warranted. due to recovery-related events (1–5). Ideal recoveries occur when a Although the cardiovascular effects of desflurane were not the calm, comfortable horse stands on its first attempt in a coordinated focus of this study, cardiovascular parameters were monitored and manner. A number of pharmacologic strategies have been developed were not different between groups. Cardiovascular and respiratory to produce ideal recoveries including reduction of inhalational anes- effects of desflurane are similar to those of other volatile inhalational thetic requirement during the maintenance phase of anesthesia using anesthetics in horses (13,25,27). In dogs, desflurane has been associ- intraoperative infusions (7,23), use of rapidly eliminated drugs (20), ated with an increase in systemic vascular resistance compared to sedation in the recovery period (10,11,15,16,24), and post-anesthetic isoflurane (27,28). In humans, induction with and increasing des- intravenous anesthesia (19), but no technique has produced a con- flurane concentration during maintenance of anesthesia have been sistently ideal recovery. Recovery from desflurane anesthesia in associated with a “sympathetic storm” characterized by an increase this study was not consistently graded as excellent regardless of in heart rate and arterial blood pressure (29). Increases in desflurane post-anesthetic sedation with xylazine. A previous study reported concentration in humans result in greater increases in sympathetic that recovery after desflurane anesthesia was of short duration with tone, greater activity of the renin-angiotensin system, and higher a rapid return to consciousness and with subjective recovery scores plasma levels of arginine vasopressin compared to isoflurane (30). A ranging from fair to excellent, the quality of which was affected by surge in heart rate occurred in 2 horses during desflurane anesthesia. anesthetic dose and duration (13). Desflurane has a lower blood- Tachycardia occurred in 1 horse during the first 15 min of anesthesia gas partition coefficient than other commonly used inhalational when the end-tidal desflurane concentration was steadily rising. anesthetics and therefore has the potential to improve recovery due The heart rate in this horse increased from 34 to 52 beats/min. In to its rapid elimination, which reduces residual effects compared to another horse, heart rate increased from 39 to 86 beats/min 1.5 h inhalational anesthetics with greater blood gas partition coefficients after induction and mean arterial blood pressure also increased (25). Recovery from anesthesia may be influenced by several factors from 70 to 132 mmHg, with no apparent change in anesthetic depth including individual variability in demeanor, the presence of hypo- or surgical stimulation. The heart rate and the mean arterial blood tension during anesthesia, the type of surgical procedure, duration pressure slowly decreased over the next 15 min in this horse and did of anesthesia, external stimuli, and use of sedatives (11,13,18,26). not increase again. This may indicate that the “sympathetic storm” Regardless of whether it received xylazine or not, 1 horse remained phenomenon described in humans also occurs in horses (30). Further recumbent longer than all other horses in both treatment groups. studies measuring catecholamines are needed to confirm this. The effects of xylazine on recovery quality in horses anesthetized Horses were positioned in dorsal recumbency and breathed spon- with isoflurane or sevoflurane have been extensively investigated. taneously for the first 15 min after induction. At 15 min, PaCO2 was Matthews et al (16) determined that recoveries from sevoflurane increased in all horses indicating hypoventilation, which was cor- and sevoflurane with post-anesthetic xylazine were superior to rected by mechanical ventilation. Concentrations of partial pressure recoveries in horses anesthetized with isoflurane. In another study, of arterial oxygen (PaO2) increased in 5 of the 6 horses during both post-anesthetic xylazine administration was associated with better anesthetic events, and pH and PaCO2 had normalized in all horses recovery scores in isoflurane-anesthetized horses than with post- within 30 min of initiating ventilation. These data are consistent anesthetic xylazine/ketamine or acepromazine, but the authors with previous observations that desflurane, like other inhalational did not evaluate recoveries without post-anesthetic sedation (15). anesthetics, produces dose-dependent respiratory depression (25). Two recent studies evaluated recoveries in isoflurane-anesthetized Limitations of the study presented here are the relatively small horses that received either xylazine or romifidine (10,24). One study number of horses, a lack of standardization of the surgical stimulus reported no difference in recovery scores (10), while another study between the first and second anesthetic events, and the application reported superior recoveries with romifidine (25). Santos et al (11) of ropes in the recovery stall. This study was carried out in conjunc- reported longer recovery times and improved recovery scores in tion with another study on bone healing, which dictated the number isoflurane-anesthetized horses that received post-anesthetic xyla- of horses used. Inclusion of additional horses may have increased zine, detomidine, or romifidine compared with no sedation. In our our ability to detect differences in recovery quality. For the study study, recovery times were longer with xylazine administration on bone healing, horses were anesthetized first for osteotomies and post-anesthesia, but overall recovery score was not different between bone drilling. During the second anesthetic event, bone growth- the 2 groups. This may have been due to the differences in blood stimulating cells were injected into these sites. Although the surgical gas solubility and shorter recovery times in desflurane-anesthetized stimulus was thus different between the 2 events, surgical times were

2000;64:0–00 The Canadian Journal of Veterinary Research 107 FOR PERSONAL USE ONLY

not different. The order of sedation was randomized to minimize the during recovery from isoflurane anaesthesia in horses. Equine effects of the surgical stimulus between groups and there was no dif- Vet J 2003;35:170–175. ference in anesthesia time, vaporizer setting, or fresh gas flow rates 12. Steffey EP. Recent advances in inhalation anesthesia. Vet Clin between the 2 groups. Ropes were applied to the horses‘ heads and North Am Equine Pract 2002;18:159–168. tails in recovery due to the potential for fracture after osteotomies. 13. Steffey EP, Woliner MJ, Puschner B, Galey FD. Effects of desflu- Our results may have been different if we had not applied these rane and mode of ventilation on cardiovascular and respiratory ropes. Recovery assistants ensured that the ropes posed no risk of functions and clinicopathologic variables in horses. Am J Vet Res entanglement to the horses by allowing enough slack in each rope 2005;66:669–677. so that, even though the ropes were attached, they were not used 14. Steffey EP, Mama KR, Brosnan RJ, et al. Effect of administration to assist recovery in any of the horses. Tension would have been of propofol and xylazine hydrochloride on recovery of horses applied only in the event of injury, or the likelihood of injury, to after four hours of anesthesia with desflurane. Am J Vet Res the horse that would have been the equivalent of a grade 4 on the 2009;70:956–963. recovery scoring system. 15. Valverde A, Black B, Cribb NC, Hathway A, Daw A. Assessment Results of the study reported here indicate that xylazine sedation of unassisted recovery from repeated general isoflurane anesthe- after desflurane anesthesia resulted in longer recovery times, but sia in horses following post-anesthetic administration of xylazine there was no difference in overall quantitative recovery scores in our or acepromazine or a combination of xylazine and ketamine. Vet model with surgery. There is considerable variation in temperament Anaesth Analg 2013;40:3–12. and behavior among horses during recovery from anesthesia and 16. Matthews NS, Hartsfield SM, Mercer D, Beleau MH, recovery characteristics are likely influenced by multiple factors, MacKenthum A. Recovery from sevoflurane anesthesia in including post-anesthetic sedation. horses: Comparison to isoflurane and effect of postmedication with xylazine. Vet Surg 1998;27:480–485. References 17. Ishihara A, Zekas LJ, Litsky AS, Weisbrode SE, Bertone AL. Dermal fibroblast-mediated BMP2 therapy to accelerate bone heal- 1. Bidwell LA, Bramlage LR, Rood WA. Equine perioperative fatali- ing in an equine osteotomy model. J Orthop Res 2010;28:403–411. ties associated with general anaesthesia at a private practice — 18. Muir WW, Hubbell JAE. Considerations for induction, main- A retrospective case series. Vet Anaesth Analg 2007;34:23–30. tenance, and recovery. In: Muir WW, Hubbell JAE, eds. Equine 2. Johnston GM, Taylor PM, Holmes MA, Wood JL. Confidential Anesthesia Monitoring and Emergency Therapy. St. Louis, enquiry of perioperative equine fatalities (CEPEF-1): Preliminary Missouri: Saunders Elsevier, 2009:381–396. results. Equine Vet J 1995;27:193–200. 19. Wagner AE, Mama KR, Steffey EP, Hellyer PW. A comparison 3. Mee AM, Cripps PJ, Jones RS. A retrospective study of mortality of equine recovery characteristics after isoflurane or isoflurane associated with general anaesthesia in horses: Elective proce- followed by a xylazine-ketamine infusion. Vet Anaesth Analg dures. Vet Rec 1998;142:275–276. 2008;35:154–160. 4. Young SS, Taylor PM. Factors influencing the outcome of equine 20. Mama KR, Steffey EP, Pascoe PJ. Evaluation of propofol for anaesthesia: A review of 1,314 cases. Equine Vet J 1993;25: general anesthesia in premedicated horses. Am J Vet Res 1996; 147–152. 57:512–516. 5. Senior JM, Pinchbeck GL, Allister R, et al. Reported morbidities 21. Viera AJ, Garrett JM. Understanding interobserver agreement: following 861 anaesthetics given at four equine hospitals. Vet The kappa statistic. Fam Med 2005;37:360–363. Rec 2007;160:407–408. 22. Landis JR. The measurement of observer agreement for categori- 6. Johnston GM, Eastment JK, Wood JLN, Taylor PM. The confi- cal data. Biometrics 1977;33:159–174. dential enquiry into perioperative equine fatalities (CEPEF): 23. Valverde A, Gunkel C, Doherty TJ, Giguère S, Pollack AS. Effect Mortality results of phases 1 and 2. Vet Anaesth Analg 2002;29: of a constant rate infusion of lidocaine on the quality of recovery 159–170. from sevoflurane or isoflurane general anaesthesia in horses. 7. Yamashita K, Muir WW 3rd, Tsubakishita S, et al. Infusion of Equine Vet J 2005;37:559–564. guaifenesin, ketamine, and medetomidine in combination with 24. Woodhouse KJ, Brosnan RJ, Nguyen KQ, Moniz GW, Galuppo inhalation of sevoflurane versus inhalation of sevoflurane alone LD. Effects of postanesthetic sedation with romifidine or xyla- for anesthesia of horses. J Am Vet Med Assoc 2002;221:1150–1155. zine on quality of recovery from isoflurane anesthesia in horses. 8. Grosenbaugh D, Muir WW. Cardiorespiratory effects of sevoflu- J Am Vet Med Assoc 2013;242:533–539. rane, isoflurane, and halothane anesthesia in horses. Am J Vet 25. Santos M, López-Sanromán J, Garcia-Iturralde P, Fuente M, Res 1998;59:101–106. Tendillo FJ. Cardiopulmonary effects of desflurane in horses. 9. Hubbell JAE. Recovery from anaesthesia in horses. Equine Vet Vet Anaesth Analg 2005;32:355–359. Educ 1999;11:160–167. 26. Whitehair KJ, Steffey EP, Willits NH, Woliner MJ. Recovery 10. Bauquier SH, Kona-Boun JJ. Comparison of the effects of xyla- of horses from inhalation anesthesia. Am J Vet Res 1993;54: zine and romifidine administered perioperatively on the recov- 1693–1702. ery of anesthetized horses. Can Vet J 2011;52:987–993. 27. Clarke KW, Song DY, Alibhai HI, Lee YH. Cardiopulmonary 11. Santos M, Fuente M, Garcia-Iturralde R, Herran R, Lopez- effects of desflurane in ponies, after induction of anaesthesia Sanroman J, Tendillo FJ. Effects of alpha-2 adrenoceptor agonists with xylazine and ketamine. Vet Rec 1996;139:180–185.

108 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

28. Kersten J, Pagel PS, Tessmer JP, Roerig DL, Schmelling WT, 30. Weiskopf RB, Moore MA, Eger EI 2nd, et al. Rapid increase in Warltier DC. Dexmedetomidine alters the hemodynamic effects desflurane concentration is associated with greater transient of desflurane and isoflurane in chronically instrumented dogs. cardiovascular stimulation than with rapid increase in isoflurane Anesthesiology 1993;79:1022–1032. concentration in humans. Anesthesiology 1994;80:1035–1045. 29. Clarke K. Desflurane and sevoflurane. New volatile anesthetic agents. Vet Clin North Am Small Anim Pract 1999;29:793–810.

2000;64:0–00 The Canadian Journal of Veterinary Research 109 FOR PERSONAL USE ONLY Compte rendu Review Article

Les biofilms bactériens : leur importance en santé animale et en santé publique Yannick D.N. Tremblay, Skander Hathroubi, Mario Jacques

Résumé Les biofilms bactériens sont des amas structurés de cellules bactériennes enrobés d’une matrice polymérique et attachés à une surface. Le biofilm protège les bactéries et leur permet de survivre dans des conditions environnementales hostiles. Les bactéries du biofilm peuvent résister à la réponse immunitaire de l’hôte et sont beaucoup plus résistantes aux antibiotiques et aux désinfectants que les cellules bactériennes planctoniques. La capacité de former un biofilm est maintenant reconnue comme une caractéristique propre à plusieurs microorganismes. La présence de biofilms lors d’infections demande donc de nouvelles méthodes de prévention, de diagnostic et de traitement. De même, la présence de biofilms sur des surfaces retrouvées à la ferme, à l’abattoir ou à l’usine de transformation affectera l’efficacité du protocole de désinfection. De façon surprenante, la formation de biofilms chez les bactéries pathogènes des animaux et les bactéries zoonotiques est un sujet relativement peu étudié. Ce bref compte rendu a pour objectif de sensibiliser les intervenants en santé animale à l’importance des biofilms.

Abstract Bacterial biofilms are structured communities of bacterial cells enclosed in a self-produced polymer matrix that is attached to a surface. Biofilms protect and allow bacteria to survive and thrive in hostile environments. Bacteria within biofilms can withstand host immune responses, and are much less susceptible to antibiotics and disinfectants when compared to their planktonic counterparts. The ability to form biofilms is now considered an attribute of many microorganisms. Diseases associated with biofilms require novel methods for their prevention, diagnosis and treatment; this is largely due to the properties of biofilms. Furthermore, the presence of biofilms on surfaces found at farms, slaughterhouses or food processing plants will have an impact on the efficacy of disinfection protocols. Surprisingly, biofilm formation by bacterial pathogens of veterinary or zoonotic importance has received relatively little attention. The objective of this brief Review article is to bring awareness about the importance of biofilms to animal health stakeholders. (Translated by the authors)

Caractéristiques des biofilms développement du biofilm survient en réponse à des signaux extracellulaires, soit présents dans l’environnement ou produits par Les biofilms bactériens sont généralement définis comme des les cellules bactériennes; (iii) le biofilm protège les bactéries contre agrégats de cellules bactériennes attachés à une surface et enrobés le système immunitaire de l’hôte, la dessiccation et les biocides d’une matrice polymérique (1–3) (Figure 1). Les bactéries peuvent (antibiotiques et désinfectants). Cette dernière caractéristique est tout aussi bien adhérer à une surface biotique (e.g. cellules de la particulièrement importante. La présence de biofilms pourra avoir muqueuse) qu’à une surface abiotique (e.g. plancher ou équipement comme effet d’interférer avec un traitement adéquat de l’animal ou à la ferme, à l’abattoir ou à l’usine de transformation). La formation une désinfection efficace des surfaces (1–6). d’un biofilm se fait en plusieurs étapes selon un modèle bien établi (Figure 2). La capacité de former un biofilm est maintenant reconnue Étapes de la formation comme une caractéristique propre à plusieurs microorganismes. On estime d’ailleurs que 80 % de la biomasse microbienne de notre d’un biofilm planète réside sous forme d’un biofilm (4). Tel que mentionné précédemment, la formation d’un biofilm Les biofilms bactériens isolés de divers environnements partagent est constituée de plusieurs étapes (Figure 2) (1–3). Le biofilm peut des caractéristiques communes (1–3) : (i) les cellules bactériennes se former très rapidement, en quelques heures; un bon exemple sont retenues ensemble par une matrice polymérique composée étant la plaque dentaire. Les bactéries doivent, dans un premier d’exopolysaccharides, de protéines et d’acides nucléiques; (ii) le temps, adhérer à une surface biotique ou abiotique. Cette étape

Réseau canadien de recherche sur la mammite bovine et la qualité du lait (Tremblay, Jacques); Centre de recherche en infectiologie porcine (Tremblay, Hathroubi, Jacques); Faculté de médecine vétérinaire, Université de Montréal, St-Hyacinthe, Québec, Canada J2S 7C6. Adresser toute correspondance à Docteur Mario Jacques; téléphone : 450-773-8521 (poste 8348); fax : 450-778-8108; e-mail : [email protected] L’adresse actuelle de Dr Jacques est Faculté de médecine vétérinaire, Université de Montréal, 3200, rue Sicotte, St-Hyacinthe, Québec, Canada J2S 7C6. Ce texte est dédié à la mémoire de J. William (Bill) Costerton; un ami et un mentor.

110 The Canadian Journal of Veterinary Research 2014;78:000–00014;78:110–116 FOR PERSONAL USE ONLY

requiert généralement la présence de molécules ou de structures particulières à la surface de la bactérie (e.g. fimbriae, flagelle). Puis les cellules bactériennes vont s’agglutiner, se multiplier et former des microcolonies. Lors de l’étape de maturation du biofilm, les bactéries synthétisent un exopolysaccharide et d’autres constituants de la matrice polymérique. Le biofilm mature représente une structure complexe et les bactéries de diverses régions du biofilm pourront exprimer des gènes différents. L’étape finale de la formation d’un biofilm est le détachement et la dispersion de cellules bactériennes (7). Ces cellules ont la capacité d’adhérer à de nouvelles surfaces et de reformer un biofilm. Le détachement des cellules peut être initié par différents facteurs : des perturbations mécaniques (e.g. force de cisaillement, abrasion), la dégradation enzymatique de la matrice polymérique (e.g. dispersine B), la dégradation enzymatique du substrat sur lequel le biofilm est attaché (e.g. hyaluronidase), l’induction de la motilité, la production d’agents tensioactifs (e.g. rhamnolipides) et le relâchement de l’exopolysaccharide. Il est bien connu que les bactéries réagissent à de multiples signaux environnementaux (e.g. concentration d’éléments nutritifs) ainsi qu’à la densité de cellules bactériennes, phénomène mieux connu sous le nom de quorum sensing, et que ceci aura une influence sur les différentes étapes de la formation du biofilm (8). Le Figure 1. Biofilm produit par une souche de Staphylococcus epidermidis détachement et la dispersion de cellules bactériennes d’un biofilm dans un système en microplaque. Image obtenue en microscopie confocale TM ® jouent un rôle important dans la transmission de bactéries de suite à une coloration du biofilm à l’aide de FilmTracer FM 1-43. réservoirs environnementaux à un hôte (animal ou humain), dans la transmission entre les hôtes et dans la propagation de l’infection chez un hôte (7). les bactéries d’un biofilm sont généralement moins sensibles aux antibiotiques et aux désinfectants que ces mêmes bactéries sous Composition de la matrice forme planctonique. En fait, les bactéries d’un biofilm peuvent être de 10 à 1000 fois plus résistantes aux agents antimicrobiens (10, 11). du biofilm Plusieurs facteurs peuvent expliquer la plus grande résistance (certains auteurs préfèrent plutôt parler d’une tolérance) des biofilms La matrice du biofilm est hautement hydratée et peut contenir aux agents antimicrobiens (2, 5, 11). Un de ces facteurs est la matrice jusqu’à 97 % d’eau. Elle peut être constituée de polysaccharides, de polymérique qui agit comme barrière réduisant ou empêchant la protéines, d’acides nucléiques, d’agents tensioactifs, de lipides, de diffusion des agents antimicrobiens. Les charges électrostatiques à la glycolipides et de cations (8, 9). La composition de la matrice varie surface de la matrice polymérique peuvent aussi lier certains agents selon l’espèce bactérienne et les conditions de croissance. Un des antimicrobiens. Le métabolisme des bactéries d’un biofilm joue exopolysaccharides le plus souvent retrouvés est un polymère de également un rôle très important. Étant donné la faible concentration b-1,6-N-acétyl-D-glucosamine (polyglucosamine, PGA ou PNAG). de certains nutriments et le gradient en oxygène, certaines cellules On retrouve ce polymère chez plusieurs espèces bactériennes, dont du biofilm seront peu actives métaboliquement et pourront même Escherichia coli, Staphylococcus aureus et Staphylococcus epidermidis, être sous forme dormante; ces cellules bactériennes dormantes Yersinia pestis, Bordetella spp. et Actinobacillus spp. La cellulose, un sont d’ailleurs probablement responsables d’une grande partie de polymère linéaire de glucose, est également retrouvée fréquemment la tolérance associée aux biofilms (12). La proximité spatiale des chez diverses espèces et genres bactériens, dont E. coli, Salmonella, bactéries au sein d’un biofilm mature favorise probablement le Enterobacter et Pseudomonas. Il existe plusieurs autres polysaccharides transfert horizontal de gènes et l’augmentation de la résistance aux dont le glucane chez Streptococcus mutans et l’alginate retrouvé chez antibiotiques. Pseudomonas. La concentration minimale inhibitrice (CMI) demeure la mesure de référence pour déterminer la sensibilité des bactéries Résistance des biofilms à planctoniques à divers agents antimicrobiens. Cependant, il faut divers agents antimicrobiens être conscient qu’il n’existe pas une bonne corrélation entre la CMI de cultures bactériennes planctoniques et la concentration Les biofilms retrouvés dans des environnements naturels ou de l’agent antimicrobien requise pour éradiquer un biofilm (10). Il industriels sont résistants aux bactériophages, aux amibes ainsi est maintenant possible de déterminer la concentration minimale qu’aux biocides utilisés dans les procédés industriels (1). Comme d’un antibiotique pouvant éradiquer un biofilm (ou «minimal indiqué précédemment, les biofilms permettent aux bactéries biofilm eradication concentration», MBEC). En effet, un groupe de pathogènes de résister à la réponse immunitaire de l’hôte. De plus, chercheurs de l’University of Calgary a développé une technologie

2000;64:0–00 The Canadian Journal of Veterinary Research 111 FOR PERSONAL USE ONLY

Figure 2. Étapes de la formation et de la dispersion d’un biofilm bactérien.

permettant de déterminer la sensibilité de biofilms à divers d’un biofilm à l’interface solide-liquide en présence d’une force antibiotiques et désinfectants (13). L’appareil originalement connu de cisaillement et d’un apport continu d’éléments nutritifs. Ce sous le nom de «Calgary Biofilm Device» est maintenant appelé système reproduit assez bien les conditions auxquelles sont exposés «MBEC Assay» et est disponible commercialement (Innovotech inc., les biofilms dans leur environnement naturel. De plus, il permet Edmonton, AB). Son format de 96 puits, bien que ne reproduisant de visualiser facilement le biofilm sous un microscope confocal. évidemment pas à la perfection l’environnement normal d’un biofilm Une modification apportée au système permet la formation d’un ou d’un biofilm formé dans une cuve à flux continu, permet le biofilm à l’interface air-liquide en présence d’une faible force de criblage rapide d’agents antimicrobiens (seuls ou en combinaison). cisaillement (15). Ce système (appelé «drip flow reactor») serait plus Ces chercheurs ont été les premiers à démontrer une augmentation représentatif de l’environnement des poumons et de la cavité orale. de la résistance aux antibiotiques lors de la croissance sous forme Plus récemment, des systèmes de microfluidique ont été développés d’un biofilm d’une vaste gamme de bactéries pathogènes des et adaptés à l’étude des biofilms (16). Les biofilms peuvent également animaux (10). être étudiés dans un système statique en microplaques de 96 puits (17). Ce système très populaire permet la formation d’un biofilm Modèles de laboratoire pour à l’interface solide-liquide et le criblage à haut débit. Le Calgary Biofilm Device, mentionné à la section précédente, combine les l’étude des biofilms caractéristiques d’un système de microplaques à 96 puits auquel Plusieurs systèmes sont couramment utilisés afin d’étudier en une force de cisaillement constante est appliquée (12). Finalement, la laboratoire la formation de biofilms (14). Les cuves à flux continu pellicule qui se forme à l’interface air-liquide d’une culture en tube (ou «flow cell») sont de petites chambres qui permettent la formation représente un autre modèle d’étude des biofilms.

112 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

Molécules pouvant interférer S. epidermidis et E. coli (34). La DNAse quant à elle peut hydrolyser les acides désoxyribonucléiques contenus dans la matrice de certains avec la formation du biofilm biofilms (35, 36). Les bactériophages produisent certaines enzymes capables de dégrader les polysaccharides de la matrice du biofilm Il existe différentes stratégies permettant d’inhiber la formation et provoquer sa dispersion (37). La spécificité des bactériophages du biofilm (4, 18–22). Celles-ci peuvent prévenir l’adhérence initiale pour un groupe de bactéries bien précis suggère par contre qu’un du microorganisme, prévenir la croissance microbienne, empêcher cocktail composé de plusieurs bactériophages soit nécessaire afin la communication entre les cellules bactériennes, inhiber la synthèse d’être efficace. de la matrice polymérique ou bien dégrader cette matrice. Le Finalement, l’interférence avec des signaux bactériens représentent criblage à haut débit de banques de composés chimiques (naturels une approche privilégiée par plusieurs chercheurs. Par exemple, la ou synthétiques) est utilisé afin d’identifier des produits capables furanone (molécule extraite d’une algue marine) et des molécules d’interférer avec la formation d’un biofilm ou capables de disperser analogues sont capables d’inhiber le quorum sensing et donc un biofilm existant (4, 18). Bien que la plupart de ces produits d’interférer par la même occasion avec la formation/dispersion antibiofilm ne tuent pas les bactéries, elles peuvent toutefois les du biofilm (38, 39). Le sulfathiazole est un antibiotique pouvant rendre plus susceptibles à l’action des agents antimicrobiens ou à la interférer avec la biosynthèse de la molécule signal c-di-GMP jouant réponse immunitaire de l’hôte. La grande majorité de ces produits un rôle important dans la formation/dispersion de biofilm chez est toujours au stade expérimental. Nous donnerons ici quelques plusieurs espèces bactériennes (40). exemples démontrant la diversité des molécules antibiofilm et de leurs cibles potentiels. Davies et Marques (23) ont identifié une petite molécule signal Formation de biofilm chez (acide cis-2-décénoïque) produite par Pseudomonas aeruginosa et les bactéries pathogènes capable d’inhiber le développement d’un biofilm et d’induire sa dispersion. Cette molécule s’est également montrée efficace contre des animaux des biofilms produits par d’autres bactéries à Gram-négatif, à Gram- Plusieurs espèces et genres bactériens causant des infections positif et par des levures. Certains acides aminés D (dextrogyres) chez l’animal, et ayant ou non un potentiel zoonotique, sont produits par les bactéries (e.g. Bacillus) sont capables de disperser les capables de former des biofilms (Tableau 1). L’espace limité nous biofilms produits par différentes bactéries à Gram-positif et à Gram- empêche d’inclure une liste exhaustive de références sur ces espèces négatif (24). Il a été récemment rapporté que certains polysaccharides bactériennes mais le lecteur intéressé est invité à consulter un article bactériens avaient la capacité d’inhiber la formation de biofilms chez de synthèse récent sur le sujet (41). Mentionnons à titre d’exemple un vaste spectre de bactéries et de champignons (25). que S. aureus est souvent impliqué dans les cas de mammite chez La bromoageliferine, molécule d’origine marine de la famille la vache laitière, une maladie économiquement très importante. Il des 2-amino-imidazoles, a démontré une activité antibiofilm chez est suggéré par plusieurs chercheurs que la formation d’un biofilm P. aeruginosa et d’autres bactéries à Gram-négatif et à Gram-positif favorise la persistance dans la glande mammaire tout en diminuant (26). Une molécule synthétique, le 2-amino-imidazole/triazole l’efficacité des traitements aux antibiotiques (42). Plusieurs gènes (ou 2-AIT), a démontré une activité antibiofilm chez les bactéries impliqués dans la formation du biofilm ont été identifiés dont ceux à Gram-négatif et à Gram-positif ainsi qu’une activité synergique (gènes icaADBC) codant pour la synthèse de l’exopolysaccharide avec des antibiotiques traditionnels (27). Truchado et coll. (28) PGA de la matrice polymérique. La capacité de former un biofilm a ont démontré que le miel de châtaignier contient un inhibiteur du également été observée chez d’autres espèces pouvant causer une quorum sensing pouvant réduire la formation d’un biofilm chez mammite chez la vache laitière, dont S. epidermidis (un staphylocoque diverses bactéries à Gram-négatif. Une approche chémoinformatique à coagulase négative ou SCN) et Streptococcus agalactiae. Notre a permis à Dürig et coll. (29) d’identifier des produits naturels laboratoire s’intéresse d’ailleurs à la formation de biofilms par les d’origine végétale (acide ellagique, esculétine et fisétine) inhibant SCN isolés du lait provenant de fermes canadiennes (Figure 1) (43). la formation de biofilm chez S. aureus et Streptococcus dysgalactiae. Fait intéressant, plus les souches proviennent d’échantillons de lait Les peptides cationiques antimicrobiens peuvent également affecter prélevés tardivement au cours de la lactation plus elles ont tendance la formation du biofilm. Par exemple, le peptide LL-37, membre à démontrer une grande capacité à former des biofilms. Nous de la famille des cathélicidines retrouvées chez les vertébrés, peut avons également observé que la croissance sous forme d’un biofilm inhiber la formation d’un biofilm chez Francisella novicida (30). diminue de façon marquée (de 4 à 500 fois) l’efficacité d’antibiotiques Notre laboratoire a démontré que de faibles concentrations de zinc d’usage courant (pénicilline G/novobiocine et ceftiofur) chez

(e.g. ZnCl2) pouvaient inhiber la formation de biofilm chez diverses les souches de CNS que nous avons testées (Tremblay YDN et bactéries pathogènes du porc (31, 32). M Jacques, données non-publiées). Plusieurs enzymes se sont montrées efficaces à dégrader la Plusieurs espèces bactériennes intéressent particulièrement matrice polymérique du biofilm. Ainsi, l’alginate lyase est capable l’industrie agroalimentaire à cause de leur capacité à causer des de dégrader l’alginate contenu dans la matrice du biofilm de infections ou des toxi-infections alimentaires chez l’homme (6, 44). P. aeruginosa (33). La dispersine B est une enzyme bactérienne C’est le cas notamment des souches entérohémorragiques d’E. coli capable de dégrader le polymère de N-acétyl-glucosamine de la O157:H7 dont le bœuf est le principal réservoir. Plusieurs de matrice du biofilm de plusieurs espèces bactériennes dont S. aureus, ces souches sont capables de former des biofilms et le plasmide

2000;64:0–00 The Canadian Journal of Veterinary Research 113 FOR PERSONAL USE ONLY

Tableau 1. Espèces et genres bactériens causant des infections diminuer le nombre de cellules viables d’un biofilm, les cellules chez l’animal, ayant ou non un potentiel zoonotique, capables viables restantes représentent une source problématique de de former des biofilms [adapté (41)] contamination. L’utilisation du Calgary Biofilm Device a permis de démontrer une plus grande résistance des salmonelles face Actinobacillus pleuropneumoniae Francisella novicida à plusieurs antibiotiques lors d’une croissance sous forme d’un Aeromonas hydrophila Francisella tularensis biofilm (10). Trueperella pyogenes groupe Bacillus cereus Bartonella henselae Haemophilus parasuis Bordetella bronchiseptica Histophilus somni Conclusions et perspectives Bordetella parapertussis Leptospira Le biofilm semble faciliter la survie des bactéries pathogènes Brucella melitensis Listeria monocytogenes dans l’environnement et chez leurs hôtes. Parce que les bactéries Burkholderia pseudomallei Mannheimia haemolytica formant des biofilms ont des caractéristiques différentes des bactéries Campylobacter coli Mycobacterium planctoniques, de nouveaux outils et de nouvelles approches Campylobacter jejuni Mycoplasma pour la prévention, le traitement et le diagnostic de ces bactéries Clostridium perfringens Pasteurella multocida pathogènes sont requis. L’impact des biofilms en santé animale Corynebacterium pseudotuberculosis Pseudomonas aeruginosa et en santé publique est très évident. La formation d’un biofilm Corynebacterium renale Riemerella anatipestifer peut augmenter la résistance aux antibiotiques, aux désinfectants Enterococcus faecalis Salmonella et à la réponse immunitaire de l’hôte. Le cas échéant, ceci aura Enterococcus faecium Staphylococcus comme effet d’interférer avec un traitement adéquat de l’animal Erysipelothrix rhusiopathiae Streptococcus ou une désinfection efficace à la ferme à l’abattoir ou à l’usine de Escherichia coli Yersinia transformation. La vaccination contre les bactéries produisant des biofilms n’a pas été évaluée de façon approfondie, mais des résultats prometteurs ont été obtenus dans certains cas. Par exemple, un vaccin composé pO157 semble y jouer un rôle essentiel (45). De plus, des souches de cellules inactivées d’une souche de S. aureus produisant un d’E. coli O157:H7 incapables de former un biofilm peuvent tout abondant biofilm induit une forte production d’anticorps contre de même s’associer à un biofilm produit par une souche non le PGA et protège contre la mammite ovine et bovine (58, 59). De pathogène d’E. coli et profiter ainsi de la protection fournie par le même, un vaccin composé de cellules d’Aeromonas hydrophila sous biofilm (46). forme d’un biofilm induit une réponse immunitaire protectrice chez Listeria monocytogenes, agent de la listériose, est un microorganisme le poisson (60). pouvant survivre et croître dans diverses conditions environnementales Le développement de la prochaine génération d’agents incluant la température de réfrigération, un pH acide et une haute antimicrobiens devra tenir compte de la synergie possible entre concentration en sel (47). Il peut être isolé d’aliments crus ou deux antibiotiques, ou entre un antibiotique et une autre molécule transformés (ex. lait, produits laitiers, viandes, poissons et fruits (ex. biocide, ion métallique, enzyme, agent tensioactif ou un de mer). La formation d’un biofilm par L. monocytogenes et le fait inhibiteur du quorum sensing) afin d’accroitre leur efficacité envers que L. monocytogenes peut également se retrouver dans des biofilms les biofilms (11). Pour ce faire, il faudra également développer des multi-espèces (48) compliquent par le fait même les procédures essais standardisés permettant d’évaluer ces synergies potentielles. de désinfection. Il a été montré que l’hydrophobicité de la surface De même, l’efficacité des biocides/désinfectants devra être évaluée bactérienne (49) et la motilité due aux flagelles (50) sont tous les deux contre des bactéries d’un biofilm et non plus uniquement contre des importants dans le développement des biofilms chez cette espèce. cellules planctoniques. Considérant sont impact sur la santé animale L’ADN semble être un composant important de la matrice du biofilm et la santé publique, la persistance de gènes de résistance aux (51). antibiotiques à l’intérieur des biofilms est un autre aspect important Les Campylobacter thermophiles, C. jejuni et C. coli, sont responsables qui ne peut être négligé. De plus, les bactéries ne forment que très d’un grand nombre d’infections gastro-intestinales. Ces bactéries rarement des biofilms composés d’une seule espèce bactérienne. peuvent former des biofilms et survivre sur les surfaces d’équipement Certaines bactéries peuvent aussi s’incorporer à des biofilms des usines de transformation (52). Ces bactéries peuvent également existants. Il existe d’ailleurs des exemples bien documentés chez former des biofilms dans les systèmes d’approvisionnement en eau à L. monocytogenes et E. coli O157:H7. L’existence de biofilms multi- la ferme (53). Ces biofilms peuvent représenter une source continue espèces amène d’ailleurs un niveau de complexité supplémentaire de contamination des animaux. Ces espèces bactériennes peuvent à l’étude et au contrôle des biofilms. également s’incorporer à des biofilms existants (54). Les flagelles En résumé, il nous faut effectuer des recherches en vue de (53, 55) et le quorum sensing (53) semblent requis pour une formation développer des stratégies pour la prévention et le traitement des optimale du biofilm. infections chez l’animal en tenant compte des caractéristiques du Les salmonelles sont également capables de former des biofilms biofilm. Des recherches sont également requises afin de développer sur une variété de surfaces (56). Les salmonelles d’un biofilm sont des procédures de désinfection permettant d’éliminer les biofilms à plus résistantes à l’action des désinfectants que les salmonelles la ferme, à l’abattoir ou à l’usine de transformation, car ces biofilms planctoniques (57). Bien que les désinfectants puissent grandement représentent des réservoirs potentiels d’agents infectieux.

114 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

Remerciements 18. Landini P, Antoniani D, Burgess JG, Nijland R. Molecular mechanisms of compounds affecting bacterial biofilm formation Les travaux sur les biofilms réalisés dans le laboratoire de MJ sont and dispersal. Appl Microbiol Biotechnol 2010;86:813–823. financés par le Conseil de recherches en sciences naturelles et en génie 19. Rendueles O, Ghigo JM. Multi-species biofilms: How to avoid (CRNSG), le Fonds de recherche du Québec — Nature et technologies unfriendly neighbors. FEMS Microbiol Rev 2012;36:972–989. (FRQ-NT), le Conseil canadien de la santé porcine et la Grappe de 20. Lebeaux D, Ghigo JM. Infections associées aux biofilms — recherche laitière du Canada. Les auteurs tiennent à remercier les Quelles perspectives thérapeutiques issues de la recherche deux évaluateurs anonymes pour la pertinence de leurs commentaires. fondamentale? Med Sci (Paris) 2012;28:727–739. 21. Worthington RJ, Richards JJ, Melander C. Small molecule control Références of bacterial biofilms. Org Biomol Chem 2012;10:7457–7474. 22. Yang L, Liu Y, Wu H, et al. Combating biofilms. FEMS Immunol 1. Costerton JW, Stewart PS, Greenberg EP. Bacterial biofilms: Med Microbiol 2012;65:146–157. A common cause of persistent infections. Science 1999;284: 23. Davies DG, Marques CNH. A fatty acid messenger is responsible 1318–1322. for inducing dispersion in microbial biofilms. J Bacteriol 2. Hall-Stoodley L, Stoodley P. Evolving concepts in biofilm 2009;191:1393–1403. infections. Cell Microbiol 2009;11:1034–1043. 24. Kolodkin-Gal I, Romero D, Cao S, Clardy J, Kolter R, Losick R. 3. Hall-Stoodley L, Costerton JW, Stoodley P. Bacterial biofilms: D-amino acids trigger biofilm disassembly. Science 2010;328: From the natural environment to infectious diseases. Nat Rev 627–629. Microbiol 2004;2:95–108. 25. Rendueles O, Kaplan JB, Ghigo JM. Antibiofilm polysaccharides. 4. Richards JJ, Melander C. Controlling bacterial biofilms. Environ Microbiol 2012;doi: 10.1111/j.1462-2920.2012.02810.x ChemBioChem 2009;10:2287–2294. 26. Huigens RW III, Ma L, Gambino C, et al. Control of bacterial 5. Anderson GG, O’Toole GA. Innate and induced resistance biofilms with marine alkaloid derivatives. Mol BioSyst 2008;4: mechanisms of bacterial biofilms. Curr Top Microbiol Immunol 614–621. 2008;322:85–105. 27. Rogers SA, Huigens RW III, Cavanagh J, Melander C. Synergistic 6. Van Houdt R, Michiels CW. Biofilm formation and the food effects between conventional antibiotics and 2-aminoimidazole- industry, a focus on the bacterial outer surface. J Appl Microbiol derived antibiofilm agents. Antimicrob Agents Chemother 2010;109:1117–1131. 2010;54:2112–2118. 7. Kaplan JB. Biofilm dispersal: Mechanisms, clinical implications, 28. Truchado P, Gil-Izquierdo A, Tomás-Barberán F, Allende A. and potential therapeutic uses. J Dent Res 2010;89:205–218. Inhibition by chestnut honey of N-acyl-L-homoserine lactones 8. Karatan E, Watnick P. Signals, regulatory networks, and materials and biofilm formation in Erwinia carotovora, Yersinia enterocolitica, that build and break bacterial biofilms. Microbiol Mol Biol Rev and Aeromonas hydrophila. J Agric Food Chem 2009;57:11186–11193. 2009;73:310–347. 29. Dürig A, Kouskoumvekaki I, Vejborg RM, Klemm P. 9. Flemming H-C, Wingender J. The biofilm matrix. Nat Rev Chemoinformatics-assisted development of new antibiofilm Microbiol 2010;8:623–633. compounds. Appl Microbiol Biotechnol 2010;87:309–317. 10. Olson ME, Ceri H, Morck DW, Buret AG, Read RR. Biofilm 30. Amer LS, Bishop BM, van Hoek ML. Antimicrobial and bacteria: Formation and comparative susceptibility to antibiotics. antibiofilm activity of cathelicidins and short, synthetic peptides Can J Vet Res 2002;66:86–92. against Francisella. Biochem Biophys Res Comm 2010;396: 11. Ceri H, Olson ME, Turner RJ. Needed, new paradigms in 246–251. antibiotic development. Expert Opin Pharmacother 2010;11: 31. Labrie J, Pelletier-Jacques G, Deslandes V, et al. Effects of 1233–1237. growth conditions on biofilm formation by Actinobacillus 12. Lewis K. Multidrug tolerance of biofilms and persister cells. Curr pleuropneumoniae. Vet Res 2010;41:03. Top Microbiol Immunol 2008;322:107–131. 32. Wu C, Labrie J, Tremblay YDN, Haine D, Mourez M, Jacques M. 13. Harrison JJ, Stremick CA, Turner RJ, Allan ND, Olson ME, Zinc as an agent for the prevention of biofilm formation by Ceri H. Microtiter susceptibility testing of microbes growing pathogenic bacteria. J Appl Microbiol 2013;115:30–40. on peg lids: A miniaturized biofilm model for high-throughput 33. Boyd A, Chakrabarty AM. Role of alginate lyase in cell screening. Nat Protoc 2010;5:1236–1254. detachment of Pseudomonas aeruginosa. Appl Environ Microbiol 14. Coenye T, Nelis HJ. In vitro and in vivo model systems to study 1994;60:2355–2359. microbial biofilm formation. J Microbiol Methods 2010;83:89–105. 34. Kerrigan JE, Ragunath C, Kandra L, et al. Modeling and 15. Goeres DM, Hamilton MA, Beck NA, et al. A method for growing biochemical analysis of the activity of antibiofilm agent a biofilm under low shear at the air-liquid interface using the Dispersin B. Acta Biol Hung 2008;59:439–451. drip flow biofilm reactor. Nat Protoc 2009;4:783–788. 35. Whitchurch CB, Tolker-Nielsen T, Ragas PC, Mattick JS. 16. Benoit MR, Conant CG, Ionescu-Zanetti C, Schwartz M, Matin A. Extracellular DNA required for bacterial biofilm formation. New device for high-throughput viability screening of flow Science 2002;295:1487. biofilms. Appl Environ Microbiol 2010;76:4136–4142. 36. Mann EE, Rice KC, Boles BR, et al. Modulation of eDNA release 17. Merritt JH, Kadouri DE, O’Toole GA. Growing and analyzing and degradation affects Staphylococcus aureus biofilm maturation. static biofilms. Curr Protoc Microbiol 2005;00:1B.1.1-1B.1.17. PLoS One 2009;4:e5822.

2000;64:0–00 The Canadian Journal of Veterinary Research 115 FOR PERSONAL USE ONLY

37. Donlan RM. Preventing biofilms of clinically relevant organisms formation on the surface of polyvinyl chloride. Lett Appl Microbiol using bacteriophage. Trends Microbiol 2009;17:66–72. 2010;50:618–625. 38. Hentzer M, Riedel K, Rasmussen TB, et al. Inhibition of 50. Todhanakasem T, Young GM. Loss of flagellum-based motility quorum sensing in Pseudomonas aeruginosa biofilm bacteria by a by Listeria monocytogenes results in formation of hyperbiofilms. halogenated furanone compound. Microbiology 2002;148:87–102. J Bacteriol 2008;190:6030–6034. 39. Müh U, Schuster M, Heim R, Singh A, Olson ER, Greenberg EP. 51. Harmsen M, Lappann M, Knochel S, Molin S. Role of extracellular Novel Pseudomonas aeruginosa quorum-sensing inhibitors DNA during biofilm formation by Listeria monocytogenes. Appl identified in an ultra-high-throughput screen. Antimicrob Agents Environ Microbiol 2010;76:2271–2279. Chemother 2006;50:3674–3679. 52. Peyrat MB, Soumet C, Maris P, Sanders P. Recovery of 40. Antoniani D, Bocci P, Maciag A, Raffaelli N, Landini P. Campylobacter jejuni from surfaces of poultry slaughterhouses Monitoring of diguanylate cyclase activity and of cyclic-di-GMP after cleaning and disinfection procedures: Analysis of a biosynthesis by whole-cell assays suitable for high-throughput potential source of carcass contamination. Int J Food Microbiol screening of biofilm inhibitors. Appl Microbiol Biotechnol 2008;124:188–194. 2010;85:1095–1104. 53. Reeser RJ, Medler RT, Billington SJ, Jost BH, Joens LA. 41. Jacques M, Aragon V, Tremblay YDN. Biofilm formation in Characterization of Campylobacter jejuni biofilms under defined bacterial pathogens of veterinary importance. Anim Health Res growth conditions. Appl Environ Microbiol 2007;73:1908–1913. Rev 2010;11:97–121. 54. Sulaeman S, Le Bihan G, Rossero A, Federighi M, Dé E, Tresse O. 42. Melchior MB. Bovine mastitis and biofilms. In: Percival SL, Comparison between the biofilm initiation of Campylobacter jejuni Knottenbelt DC, Cochrane CA, eds. Springer Series on Biofilms, and Campylobacter coli strains to an inert surface using BioFilm Vol. 6 Biofilms and Veterinary Medicine. New York: Springer, Ring Test. J Appl Microbiol 2010;108:1303–1312. 2011:205–222. 55. Moe KK, Mimura J, Ohnishi T, et al. The mode of biofilm 43. Tremblay YDN, Lamarche D, Chever P, Haine D, Messier S, formation on smooth surfaces by Campylobacter jejuni. J Vet Med Jacques M. Characterization of the ability of coagulase-negative Sci 2010;72:411–416. staphylococci isolated from the milk of Canadian farms to form 56. Jain S, Chen J. Attachment and biofilm formation by various biofilms. J Dairy Sci 2013;96:234–246. serotypes of Salmonella as influenced by cellulose production 44. Jacques M, Tremblay YDN. Les biofilms : s’en préoccupe-t-on and thin aggregative fimbriae biosynthesis. J Food Prot 2007;70: assez dans l’industrie agroalimentaire? Centre québécois de 2473–2479. valorisation des biotechnologies, Fiche BioTendance® 2010:BTD 57. Wong HS, Townsend KM, Fenwick SG, Trengove RD, 10-7, www.cqvb.ca. O’Handley RM. Comparative susceptibility of planktonic and 45. Puttamreddy S, Cornick NA, Minion FC. Genome-wide 3-day-old Salmonella Typhimurium biofilms to disinfectants. transposon mutagenesis reveals a role for pO157 genes in biofilm J Appl Microbiol 2010;108:2222–2228. development in Escherichia coli O157:H7 EDL933. Infect Immun 58. Pérez MM, Prenafeta A, Valle J, et al. Protection from Staphylococcus 2010;78:2377–2384. aureus mastitis associated with poly-N-acetyl b-1,6 glucosamine 46. Uhlich GA, Rogers DP, Mosier DA. Escherichia coli serotype specific antibody production using biofilm-embedded bacteria. O157:H7 retention on solid surfaces and peroxide resistance is Vaccine 2009;27:2379–2386. enhanced by dual-strain biofilm formation. Foodborne Pathog 59. Prenafeta A, March R, Foix A, Casals I, Costa L. Study of Dis 2010;7:935–943. the humoral immunological responses after vaccination with 47. Gandhi M, Chikindas ML. Listeria: A foodborne pathogen that a Staphylococcus aureus biofilm-embedded bacterin in dairy knows how to survive. Int J Food Microbiol 2007;113:1–15. cows: possible role of the exopolysaccharide specific antibody 48. Habimana O, Meyrand M, Meylheuc T, Kulakauskas S, production in the protection from Staphylococcus aureus induced Briandet R. Genetic features of resident biofilms determine mastitis. Vet Immunol Immunopathol 2010;134:208–217. attachment of Listeria monocytogenes. Appl Environ Microbiol 60. Nayak DK, Asha A, Shankar KM, Mohan CV. Evaluation of 2009;75:7814–7821. biofilm of Aeromonas hydrophila for oral vaccination of Clarias 49. Takahashi H, Suda T, Tanaka Y, Kimura B. Cellular hydrophobicity batrachus — A carnivore model. Fish Shellfish Immunol 2004;16: of Listeria monocytogenes involves initial attachment and biofilm 613–619.

116 The Canadian Journal of Veterinary Research 2000;64:0–00 Article

Immunogenicity and efficacy of a recombinant adenovirus expressing hemagglutinin from the H5N1 subtype of swine influenza virus in mice Yunpu Wu, Chuanling Qiao, Huanliang Yang, Yan Chen, Xiaoguang Xin, Hualan Chen

Abstract The H5N1 influenza viruses infect a range of avian species and have recently been isolated from humans and pigs. In this study we generated a replication-defective recombinant adenovirus (rAd-H5HA-EGFP) expressing the hemagglutinin (HA) gene of H5N1 A/ Swine/Fujian/1/2001 (SW/FJ/1/01) and evaluated its immunogenicity and protective efficacy in BALB/c mice. The recombinant virus induced high levels of hemagglutination inhibition (HI) antibody at a median tissue culture infective dose of 108 or 107. Compared with mice in the control groups, the mice vaccinated with rAd-H5HA-EGFP did not show apparent weight loss after challenge with either the homologous SW/FJ/1/01 or the heterologous H5N1 A/Chicken/Hunan/77/2005 (CK/HuN/77/05). Replication of the challenge virus was partially or completely inhibited, and viruses were detected at significantly lower numbers in the organs of the vaccinated mice, all of which survived the challenge with CK/HuN/77/05, whereas most of the control mice did not. These results indicate that rAd-H5HA-EGFP can provide effective immune protection from highly pathogenic H5N1 viruses in mice and is therefore a promising new candidate vaccine against H5N1 influenza in animals.

Résumé Les virus de l’influenza de type H5N1 infectent une grande variété d’espèces aviaires et ont récemment été isolés des humains et des porcs. Dans la présente étude nous avons généré un adénovirus recombinant défectueux pour la réplication (rAd-H5HA-EGFP) exprimant le gène de l’hémagglutinine (HA) du virus H5N1 A/Porc/Fujian/1/2001 (SW/FJ/1/01) et évalué son immunogénicité et son activité protectrice chez des souris BALB/c. Le virus recombinant induisit des titres élevés d’anticorps inhibant l’hémagglutination (HI) à une dose médiane infectant les cultures cellulaires de 108 ou 107. Comparativement aux souris des groupes témoins, les souris vaccinées avec rAd-H5HA-EGFP n’ont pas montré de perte de poids apparente après une infection défi avec soit le virus homologue SW/FJ/1/01 ou le virus H5N1 hétérologue A/ Poulet/Hunan/77/2005 (CK/HuN/77/05). La réplication du virus de l’infection défi fut partiellement ou complètement inhibée, et les virus furent détectés à une charge virale significativement inférieure dans les organes des souris vaccinées, qui ont toutes survécu à l’infection défi avec le virus CK/HuN/77/05, ce qui ne fut pas le cas de la majorité des souris témoins. Ces résultats indiquent que rAd-H5HA-EGFP peut fournir une protection efficace contre des virus H5N1 hautement pathogènes chez les souris et est ainsi un nouveau vaccin candidat prometteur contre l’influenza H5N1 chez les animaux. (Traduit par Docteur Serge Messier)

Introduction under experimental conditions. Pigs inoculated with H5N1 viruses showed moderate levels of virus shedding in nasal swabs, mild or Influenza A viruses are divided into 16 subtypes on the basis of no symptoms, and no virus transmission to in-contact animals (5,6). the hemagglutinin (HA) protein and 9 subtypes on the basis of the In China in 2003 and 2005, H5N1 viruses were isolated from pigs, neuraminidase protein (1). The highly pathogenic (HP) subtype of although none of the animals died or showed evidence of related influenza virus, subtype H5N1, not only causes outbreaks of influ- illness (7–9). However, taken together, these findings highlight the enza in a variety of birds but can also fatally infect humans, cats, potential risk of swine influenza to public health. and other mammals (2,3). From the end of 2003 to February 2013, The HA protein is one of the most important antigens for the HP H5N1 viruses caused 620 confirmed cases of influenza in induction of protective immunity against influenza viruses. It can humans, 367 fatal (4). Pigs are an intermediate host of influenza induce predominantly subtype-specific antibodies that can neutral- viruses and have been shown to be susceptible to HP H5N1 viruses ize infecting virus (10). Human adenovirus type 5 (Ad5) is an ideal

Animal Influenza Laboratory of the Ministry of Agriculture, State Key Laboratory of Veterinary Biotechnology, Harbin Veterinary Research Institute, Chinese Academy of Agricultural Sciences, 427 Maduan Street, Harbin 150001, People’s Republic of China. M.S. Yunpu Wu’s current address is Liaoning Provincial Center for Animal Disease Control and Prevention, Shenyang 110164, People’s Republic of China. Address all correspondence to Dr. Chuanling Qiao or Dr. Hualan Chen; telephone: 86 451 51997168; fax: 86 451 51997166; e-mail: [email protected] or [email protected] Received November 6, 2012. Accepted March 15, 2013.

2014;78:117–126 The Canadian Journal of Veterinary Research 117 FOR PERSONAL USE ONLY

expression vector for viral antigens because it is easily manipulated adenovirus backbone plasmid (pBHGloxDE1,E3Cre) at an appropri- and produced under conditions of good manufacturing practice. The ate ratio. Adenovirus vector pFG140 was used as a control in the recombinant adenovirus (rAd) can elicit robust cellular and humoral transfection experiments. Sixteen hours after transfection the cells immunity against the encoded antigen (11), and vaccination with were checked under a fluorescence microscope. When the fluores- rAd has produced effective immunity in preclinical models (12,13). cence intensity increased in the infected cells and typical cytopathic In the present study, we generated a replication-defective rAd, effects appeared, the cells were harvested for rAd screening. After rAd-H5HA-EGFP (EGFP = enhanced green fluorescent protein), further empty plaque screening, the rAd (rAd-H5HA-EGFP) was using Ad5 as the vector and the HA gene from the H5N1 subtype of stored before its use in other experiments. swine influenza virus (SIV) as the donor. To evaluate the immuno­ genicity and efficacy of the recombinant virus, we inoculated Genetic stability of rAd-H5HA-EGFP BALB/c mice with different doses. To evaluate the genetic stability of the foreign gene after inte- gration into the rAd, rAd-H5HA-EGFP was passaged 14 times in Materials and methods HEK293 cells. DNA was extracted from the rAd from the 5th to the 14th passage (10 batches) with the Wizard SV Genomic DNA Purification System (Promega, Madison, Wisconsin, USA), and the Viruses, plasmids, and cells inserted gene was detected by polymerase chain reaction (PCR) as The H5N1 SIV A/Swine/Fujian/1/2001 (SW/FJ/1/01; accession follows. number AY747617, which has been characterized previously (8), and the H5N1 avian influenza virus A/Chicken/Hunan/77/2005 Detection of HA gene expression by the rAd (CK/HuN/77/05), kindly provided by Dr. Yanbing Li, Harbin The total RNA was extracted from HEK293 cells infected with the Veterinary Research Institute, Chinese Academy of Agricultural rAd by conventional methods. Any remaining DNA contamination Sciences, Harbin, People’s Republic of China, were propagated in was removed by treatment of the total RNA with DNase I at 37°C the allantoic cavity of 10-day-old embryonated eggs from chickens for 30 min, then inactivation at 75°C for 10 min. Reverse transcrip- free of specific pathogens. The viruses were stored at −70°C before tion was carried out with RNA as the template and Oligo (dT)-18 RNA extraction or challenge studies and were titrated to determine as the primer. The reaction mixture contained 0.5 mL of Oligo (dT)- the 50% egg infective dose (EID50) by the Reed and Muench method 18 (500 mg/mL), 4.0 mL of 53 buffer, 0.5 mL of deoxynucleotide (14). The adenovirus shuttle plasmid pDC315 and the backbone triphosphate (10 mM) (Promega), and 14.5 mL of mRNA. After plasmid pBHGloxDE1,E3Cre of the Ad-Max adenovirus system incubation of the reaction mixture at 65°C for 10 min, 0.5 mL of was purchased from Microbix Biosystems (Mississauga, Ontario). reverse transcriptase M-MLV (200 U/mL) (Promega) was added and Plasmid pIRES2-EGFP (Clontech Laboratories, Mountain View, well mixed. The reaction was then continued at 37°C for 60 min. To California, USA) was kindly provided by Professor Zhigao Bu, analyze transcription of the target gene at the mRNA level, the HA Harbin Veterinary Research Institute. gene was amplified as described above. Human embryonic kidney (HEK293) cells were cultured in To examine the biologic activity of the HA protein expressed by Dulbecco’s modified Eagle’s medium containing 10% fetal bovine rAd-H5HA-EGFP, we harvested HEK293 cells infected with the serum. rAd or with the vector virus rAd-EGFP and suspended them in phosphate-­buffered saline (PBS). An equal volume of 23 sodium Construction of the rAd shuttle plasmid dodecyl sulfate (SDS) electrophoresis sample buffer (100 mM Viral RNA was extracted with an RNeasy Mini Kit (Qiagen, Tris-HCl, pH 6.8; 200 mM dithiothreitol; 4% SDS; 0.2% bromophenol Valencia, California, USA), and then cDNA encoding the HA blue; and 20% glycerol) was then added. The samples were boiled protein of SW/FJ/1/01 was reverse-transcribed from the RNA in water for 10 min for protein denaturation and then loaded onto a with the Uni12 primer (59-AGCAAAAGCAGG-39). The HA 10% polyacrylamide gel for electrophoresis. Separated proteins were gene was amplified from the cDNA with the primers HAF probed for the HA protein with serum from chickens that had been (59-CCCGTCGACATGGAGAAAATAGTGCTTCTTCTTGC-39) and infected with the H5N1 subtype of influenza virus, prepared in our HAR (59-TCCCCCGGGTTAAATGCAAATTCTGCATTGTAACG-39). laboratory, and rabbit anti-chicken IgG antibody labeled with IRDye Artificial SalI and SmaI restriction sites (underlined) permitted direct 700DX (LI-COR Biosciences, Lincoln, Nebraska, USA). Western blot cloning of the amplification product into plasmid pIRES2-EGFP, the analysis was done according to the method of Sambrook, Fritsch, resulting plasmid being designated pHA-IRES-EGFP (IRES = internal and Maniatis (15). ribosome entry site). The rAd shuttle plasmid was constructed by inserting DNA fragments containing the HA and EGFP genes, con- Vaccination and challenge of mice nected by an IRES nucleotide sequence, into pDC315. The resultant The rAd was prepared for vaccination by proliferation in HEK293 plasmid was designated pDC315-H5HA-EGFP. cells, purification, and concentration with a Vivapure AdenoPACK 100 RT kit (Sartorius AG, Hannover, Germany). The 50% tissue Generation of the rAd culture infective dose (TCID50) of the purified virus was titrated in With use of the transfection reagent Lipofectamine 2000 HEK293 cells with the Rapid Test Kit for Adenovirus Titer [VGTC (Invitrogen, Carlsbad, California, USA), HEK293 cells were cotrans- (Vector Gene Technology Company), Beijing, People’s Republic of fected with the rAd shuttle plasmid (pDC315-H5HA-EGFP) and the China], according to the manufacturer’s instructions.

118 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

A B

C D

H5HA IRES EGFP pA

Figure 1. A — Amplification by polymerase chain reaction (PCR) of the hemagglutinin (HA) gene of the H5N1 swine influenza virus (SIV) A/Swine/ Fujian/1/2001 (SW/FJ/1/01). Lane 1 — DNA marker DL2000. Lane 2 — PCR product of the HA gene. Lane 3 — negative control. Bp — base pairs. B — Identification of the recombinant plasmid pHA-IRES-EGFP. Lane 1 — DNA marker DL15 000. Lane 2 — Products of digestion of pHA-IRES-EGFP by SalI and SmaI. Lane 3 — Product of digestion of the plasmid pIRES2-EGFP by SalI. Lane 4 — PCR product of pHA-IRES-EGFP. Lane 5 — DNA marker DL2000. IRES — internal ribosome entry site; EGFP — enhanced green fluorescent protein. C — Restriction analysis of the recombinant adenovirus (rAd) shuttle plasmid pDC315-H5HA-EGFP. Lane 1 — DNA marker DL15 000. Lane 2 — Products of digestion of the plasmid by SalI and SmaI. Lane 3 — Products of digestion of the plasmid by NheI and SmaI. D — Diagrammatic representation of the rAd rAd-H5HA-EGFP and the vector virus rAd-EGFP. ITR — inverted terminal repeat.

2000;64:0–00 The Canadian Journal of Veterinary Research 119 FOR PERSONAL USE ONLY

A B

C

Lane 3: Lysates of HEK293 cells infected with Lane 2: Molecular weight protein Lane 1: Lysates of HEK293 cells infectedrAd-H5HA-EGFP with rAd-EGFP

Figure 2. A — Cytopathic effects displayed by human embryonic kidney cells (HEK293) infected by rAd-H5HA-EGFP. B — Intense fluorescence signals of rAd-H5HA-EGFP expressing the EGFP gene. C — Results of Western blot analysis of lysates of HEK293 cells infected with rAd-EGFP or rAd-H5HA- EGFP and incubated with chicken antiserum specific for H5 SIV HA that was generated by H5HA gene vaccination. Lane 1 — Lysates of HEK293 cells infected with rAd-EGFP. Lane 2 — Molecular weight protein. Lane 3 — lysates of HEK293 cells infected with rAd-H5HA-EGFP.

Six-week-old female BALB/c mice (n = 100), purchased from and the 22 mice in the other control group were inoculated intra- Beijing Vital River Experimental Animal Company, Beijing, were muscularly with 50 mL of PBS. Three weeks after the 1st vaccination housed in a clean chamber under negative pressure. The mice were the mice in the 2 rAd5-H5HA-EGFP-inoculated groups were given 8 randomly divided into 4 groups and observed for a week before another intramuscular dose of rAd5-H5HA-EGFP, 2 3 10 TCID50 7 vaccination. Two groups of mice (28 per group) were vaccinated and 2 3 10 TCID50, respectively, to boost the immunization. Three 8 intramuscularly with rAd5-H5HA-EGFP at a dose of 10 TCID50 weeks after the 2nd vaccination the mice in each group were divided 7 6 and 10 TCID50, respectively. The 22 mice in 1 control group were into 2 subgroups, which were challenged intranasally with 10 EID50 8 inoculated intramuscularly with rAd5-EGFP at a dose of 10 TCID50, of SW/FJ/1/01 and CK/HuN/77/05, respectively. The challenge

120 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

experiments were conducted in a biosafety level 3 laboratory at 5.3-kb band (Figure 1B, lane 3). The rAd shuttle plasmid pDC315- Harbin Veterinary Research Institute. H5HA-EGFP was generated by subcloning the fragments containing All the mice were monitored daily by clinical examination, and the HA and EGFP genes, connected by an IRES, into pDC315. The their body weight was measured each day after challenge until the shuttle plasmid was similarly confirmed by PCR, and double diges- end of the experiment. Four and six days after challenge 3 mice tion with SalI/SmaI or NheI/SmaI resulted in 2 fragments, at sizes of from each subgroup were euthanized, and tissues, including brain, 5.3 and 1.7 kb, respectively (Figure 1C, lanes 2 and 3, respectively). lung, kidney, and spleen, were removed for virus titration in eggs, as previously described (16). The remaining animals were kept for Generation of the rAd expressing the HA gene clinical observation and euthanized 14 d after challenge. (rAd-H5HA-EGFP) Recombinant adenovirus shuttle plasmid pDC315-H5HA-EGFP Hemagglutination inhibition (HI) assay and backbone plasmid pBHGloxDE1,E3Cre were used to cotransfect Blood was collected from all of the mice at 3 time points: 2 wk HEK293 cells. Seven days after cotransfection the HEK293 cells and 3 wk after the 1st vaccination and 3 wk after the booster vac- became swollen and rounded. The cells were then harvested and cination. Antibodies to the HA protein of influenza viruses (includ- passaged on HEK293 cells. Typical adenovirus-induced cytopathic ing SW/FJ/1/01 and CK/HuN/77/05) were assayed in the serum effects — grape-cluster-like cells (Figure 2A) — and intensive fluores- according to the procedure described in the Manual of Diagnostic cence signals (Figure 2B) were observed by fluorescence microscopy. Tests and Vaccines for Terrestrial Animals of the World Organisation The presence of the HA gene was confirmed by PCR, and expres- for Animal Health (17). Briefly, serum samples were treated with sion of the HA protein was confirmed by reverse-transcription PCR receptor-destroying enzyme, heat-inactivated, and centrifuged. (RT-PCR) and Western blot analysis. As expected, a band correspond- Supernatant samples were then serially diluted in V-shaped well ing to the HA gene was amplified by RT-PCR only when the RNA microtiter plates with an equal volume of a virus suspension contain- isolated from the HEK293 cells infected with rAd-H5HA-EGFP was ing 4 hemagglutination units of virus, and the plates were incubated used as a template. In the Western blot analysis the expressed HA at room temperature before the addition of 0.5% chicken erythro- protein was specifically recognized by chicken polyclonal antiserum cytes. The virus titer was defined as the reciprocal of the maximum to SW/FJ/1/01 from the HEK293 cells infected with rAd-H5HA- dilution at which hemagglutination was inhibited. EGFP (Figure 2C). Figure 1D is a diagrammatic representation of rAd-EGFP and rAd-H5HA-EGFP expressing the HA gene from Virus titration SW/FJ/1/01 and the EGFP gene connected by an IRES. In contrast, Virus titration in eggs was carried out using 10% (weight/volume) the HEK293 cells infected with rAd-EGFP gave negative results in clarified tissue homogenates in PBS. Different 0.1-mL dilutions of the Western blot analysis (Figure 2C). each sample were used to inoculate 10-day-old embryonated eggs via The presence of the HA gene in the recombinant virus rAd- the allantoic cavity. The allantoic fluids were tested for HA activity H5HA-EGFP was confirmed by PCR in the 5th to 14th passages after incubation at 37°C for 72 h. The titer of virus in each sample, (10 batches) in HEK293 cells, and expression of the HA protein in expressed in EID50/mL, was calculated by the method of Reed and the infected HEK293 cells of each passage was confirmed by Western Muench (14). blot analysis. Statistical analysis Antibody responses induced by rAd-H5HA-EGFP The results from each experimental group were compared by in mice analysis of variance and the t-test. A P-value of , 0.05 was consid- The immunogenicity of rAd-H5HA-EGFP was determined by ered statistically significant. measuring the HI antibody titers in serum from the vaccinated 10 mice. The TCID50 of the rAd was found to be 2.26 3 10 /mL. Two Results weeks after the 1st vaccination, HI antibody (mean titer 1:36) was detected in the 2 groups of vaccinated mice. Three weeks after the 1st vaccination the titers reached up to 1:80 in the mice inoculated Construction of the rAd shuttle plasmid 8 with 10 TCID50 of rAd-H5HA-EGFP and 1:66 in the mice vaccinated 7 The cDNA of the HA gene derived from the SW/FJ/1/01 virus with 10 TCID50. Three weeks after the booster vaccination the mean was amplified with use of the primers HAF and HAR. As expected, titers had increased again, up to 1:120 and 1:92, respectively. In the the size of the PCR products was about 1.7 kilo base pairs (kb) control mice inoculated with rAd-EGFP or PBS, HI antibody was (Figure 1A, lane 2). The resultant intermediate plasmid, pHA-IRES- not detected. EGFP, was constructed by inserting the HA fragment into plasmid pIRES2-EGFP. When PCR was carried out to amplify the HA gene Protective efficacy of rAd-H5HA-EGFP against using pHA-IRES-EGFP as the template and HAF and HAR as H5N1 influenza virus the primers, the size of the PCR products was again about 1.7 kb Three weeks after the 2nd vaccination, we challenged the mice 6 (Figure 1B, lane 4), as expected. The plasmid pHA-IRES-EGFP was with 10 EID50 of SW/FJ/1/01 or CK/HuN/77/05. On days 4 and 6 confirmed by SalI/SmaI double digestion, which resulted in 2 bands, after challenge, 3 mice from each subgroup were euthanized, and at sizes of 1.7 and 5.3 kb, respectively (Figure 1B, lane 2). The plas- brain, lung, kidney and spleen tissue were removed for virus titra- mid pIRES2-EGFP was also digested by SalI, which generated a tion. After challenge with SW/FJ/1/01 the virus was not detected in

2000;64:0–00 The Canadian Journal of Veterinary Research 121 FOR PERSONAL USE ONLY of 50 oportion TCID Pr surviving 5/5 1/5 8/8 8/8 8/8 8/8 5/5 0/5 8

1.80 0.43

0.5 — no virus was 0.5 — no virus was 0.5 0.5 0.5 0.5 0.5 0.5 Kidney , 2.00 6 , , , , , 2.25 6

1.01 0.58 1.09 1.30 0.14

3 wk after the 2nd inoculation.

6 0.5 0.5 0.5 Lung 0.58 6 4.83 6 , , , 3.25 6 1.50 6 5.88 6 ) of 10 50

1.80 0.14

6 d after challenge 0.5 0.5 0.5 0.5 0.5 0.5 ) c Spleen , 2.00 6 , , , , , 3.42 6 /g 50

of rAd5-H5HA-EGFP or were inoculated as controls with 10 7 EID 10 0.58 0.58

or 10 8 0.5 0.5 0.5 0.5 0.5 0.5 Brain , 2.83 6 , , , , , 3.17 6 ) of 10

50 0.87 0.87

0.5 0.5 0.5 0.5 0.5 0.5 Titer of virus in organ (log Kidney , 0.50 6 , , , , , 1.00 6

1.01 0 1.75 2.59 1.26 0

0.5 0.5 Lung 0.58 6 5.50 6 , , 1.67 6 1.50 6 1.33 6 4.50 6

4 d after challenge 0 1.26

0.5 0.5 0.5 0.5 0.5 0.5 Spleen , 1.50 6 , , , , , 1.33 6

0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 Brain , , , , , , , , , respectively, and the control groups were given the same doses of inocula as given previously. The challenge experiment was conducted with the with the The challenge experiment was conducted and the control groups were given same doses of inocula as previously. respectively, , 50

b 92 20 TCID 10 10 10 10 10 120 7 HI Ab Mean titer , , , , , 10

8 7 8 7

L of phosphate-buffered saline (PBS) by intramuscularly administration. Three weeks later the vaccinated mice were given booster injections of rAd5-H5HA-EGFP, L of phosphate-buffered saline (PBS) by intramuscularly administration. Three weeks later the vaccinated mice were given booster injections of rAd5-H5HA-EGFP, a

L) 8 8

and 2 3

50 L L or m m m 50 TCID 8 10 rAd-H5HA-EGFP, 10 rAd-H5HA-EGFP, PBS, 50 PBS, 50 rAd-H5HA-EGFP, 10 rAd-H5HA-EGFP, 10 rAd5-EGFP, 10 rAd-H5HA-EGFP, 10 rAd5-EGFP, rAd-H5HA-EGFP, 10 rAd-H5HA-EGFP, 3 serum. The titers are means of titers of antibody Blood was collected from the mice 3 wk after 2nd inoculation to assay hemagglutination inhibition (HI) antibody (Ab) in serum. The titers are means of Groups of 6-wk-old BALB/c mice were vaccinated with a 50% tissue culture infective dose (TCID Three mice from each group were euthanized on days 4 and 6, the organs collected for virus titration in specific-patho gen-free chicken embryos; ,

against the specific challenge virus. initial inoculum, dose (TCID Table I. Protective efficacy of the recombinant adenovirus rAd-H5HA-EGFP against challenge with highly pathogenic H5N1 viruses in mice Table Challenge virus; a b c isolated from the undiluted sample. H5N1 viruses A/Swine/Fujian/1/2001 (SW/FJ/1/01) and A/Chicken/Hunan/77/2005 (CK/HuN/77/05) at a 50% egg infective dose (EID H5N1 viruses A/Swine/Fujian/1/2001 rAd5-EGFP or 50 m 2

SW/FJ/1/01

CK/HuN/77/05

122 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

A

8 10 TCID50 rAd-H5HA-EGFP

7 10 TCID50 rAd-H5HA-EGFP

rAd-EGFP

PBS

B

8 10 TCID50 rAd-H5HA-EGFP

7 10 TCID50 rAd-H5HA-EGFP

rAd-EGFP

PBS

C

8 10 TCID50 rAd-H5HA-EGFP

7 10 TCID50 rAd-H5HA-EGFP

rAd-EGFP

PBS

Figure 3. A — Weight change (mean and standard deviation) in mice vaccinated intramuscularly with rAd-H5HA-EGFP at a 50% tissue culture infective 8 7 8 7 dose (TCID50) of 10 (diamonds) or 10 (squares) and then 3 wk later given a booster dose of 2 3 10 TCID50 or 2 3 10 TCID50, respectively, and in 8 control groups of mice inoculated intramuscularly with rAd5-EGFP at a dose of 10 TCID50 (triangles) or with 50 mL of phosphate-buffered saline (PBS; 6 6 crosses) on 2 occasions 3 wk apart, and challenged by intranasal inoculation with SW/FJ/1/01 at a 50% egg infective dose (10 EID50) of 10 3 wk after the 2nd vaccination or the inoculation of rAd5-EGFP or PBS. B — Weight change in mice vaccinated or inoculated identically but challenged by 6 intranasal inoculation with the H5N1 avian influenza virus A/Chicken/Hunan/77/2005 (CK/HuN/77/05) at an EID50 of 10 3 wk after the 2nd vaccination 6 or the inoculation of rAd5-EGFP or PBS. C — Survival rate of the mice challenged with 10 EID50 of CK/HuN/77/05 3 wk after the 2nd vaccination or the inoculation of rAd5-EGFP or PBS.

2000;64:0–00 The Canadian Journal of Veterinary Research 123 FOR PERSONAL USE ONLY

any of the organs from the mice vaccinated with rAd-H5HA-EGFP was accurately expressed by rAd-H5HA-EGFP, producing HA pro- at either dose (Table I). Viruses were detected at lower titers in the tein with good biologic activity, and the gene was highly stable after lungs alone in the mice from both control groups. After challenge continuous passage. The TCID50 of the purified rAd was determined with CK/HuN/77/05, high titers of virus were detected at both to be 2.26 3 1010/mL. 4 and 6 d after challenge in the lungs of the mice in both control The mouse is widely accepted as a relevant model of human and groups. Furthermore, the challenge virus was also isolated from the other mammalian influenza virus infections (26–28). Although H5N1 spleen and kidney at 4 and 6 d after challenge and from the brain at influenza viruses have been isolated from pigs (8,9), the pig infection 6 d. Among the rAd-H5HA-EGFP-inoculated mice the virus titers model is not well established under laboratory conditions. Therefore, in the lungs were significantly lower (P , 0.05) than those in the the immunogenicity and efficacy of rAd-H5HA-EGFP was evaluated control groups, and no virus was detected in the brain, spleen, or in the mouse model in this study. Our findings indicated that rAd- kidney. These results indicate that vaccination with rAd-H5HA-EGFP H5HA-EGFP was immunogenic, inducing high HI antibody titers, significantly inhibits the replication of homologous and heterologous and efficacious in challenge with homologous and heterologous H5N1 virus in mice. H5N1 influenza viruses. The HP H5N1 virus CK/HuN/77/05 was The rest of the mice in each group were kept for clinical observa- used as a challenge virus in this study because preliminary inves- tion for a further 2 wk after challenge. Five days after challenge with tigations had shown that it could replicate systemically in vivo and SW/FJ/1/01 the mice inoculated with rAd-EGFP or PBS started to cause death in mice. The protective efficacy of rAd-H5HA-EGFP show ruffled fur and weight loss (Figure 3A). The 2 groups of mice against virus challenge was positively correlated with the antibody vaccinated with rAd-H5HA-EGFP remained healthy and showed no levels induced by the vaccine in mice. The vaccinated mice produced weight loss after this challenge with SW/FJ/1/01. None of the mice HI antibodies against homologous H5N1 strains, but there was died during the observation period after this challenge. little or no cross-reactivity with heterologous H5N1 viruses. The The control mice showed significant weight loss beginning rAd-H5HA-EGFP vaccine offered protection against challenge with 2 d after challenge with CK/HuN/77/05 (P , 0.05). Four mice SW/FJ/1/01 or CK/HuN/77/05, preventing disease and reducing inoculated with rAd-EGFP died within 9 to 11 d after challenge, virus replication after challenge. and 5 PBS-inoculated mice died within 8 to 11 d after challenge Previously reported adenovirus-based experimental influenza vac- (Figures 3B and 3C; Table I). All of the mice that received rAd- cines have also shown protective efficacy in mouse models of influ- H5HA-EGFP remained healthy during the observation period after enza challenge (10,19,23,29). It is difficult to make direct comparisons challenge with CK/HuN/77/05; slight weight loss was observed between the results of our study and those referenced above owing within the 1st week after challenge, followed by weight gain in the to differences in experimental design, vaccine dose, vaccination 2nd week (Figure 3B). route, the challenge virus strain used, and the dose of the challenge virus. For example, adenovirus-based vaccines carrying HP avian Discussion influenza HA genes (10,23) protected BALB/c mice from challenge with lethal HP avian influenza virus. However, the vaccinated mice Since their re-emergence in 2003, influenza A (H5N1) viruses have in those studies were all challenged with lower doses of HP avian become endemic in some countries and continue to cause outbreaks influenza virus than the mice in our study: our mice vaccinated with 6 in poultry and sporadic human infections (4). Current data suggest rAd-H5HA-EGFP were challenged with 10 EID50 of SW/FJ/1/01 or that H5N1 influenza viruses can be isolated from pigs but do not CK/HuN/77/05, whereas the animals in the previous studies were 3.7 become established in swine populations (5–9). Pigs have been challenged with 10 EID50 of A/VN/1203/04. Moreover, our mice 8 7 suggested to be “mixing vessels” capable of generating reassorted were vaccinated with a lower dose of vaccine, 10 or 10 TCID50, than influenza viruses with pandemic potential (18), which suggests the the mice in the other studies. A chimpanzee-derived replication- urgent need for a vaccine to prevent swine H5N1 influenza. defective adenovirus vector of serotype SAd-V25 that expressing Inactivated whole-virus vaccines have been widely used to control a linear B-cell epitope of the ectodomain of matrix 2 (M2e) within influenza outbreaks. However, the production of this kind of vaccine variable region 1 (VR1) induced M2e-specific antibody responses is highly dependent on chicken eggs, which limits production capac- of higher magnitude and avidity than that of vectors carrying M2e ity and can create a bottleneck in the comprehensive prevention and within variable region 4 (VR4) or expressing the M2e as part of a control of H5N1 influenza pandemics (19). Several approaches have transgene product (30). Compared with an AdHu5-HA control, a been used to generate vaccines to supplement the current egg-based replication-incompetent porcine adenovirus 3 vector expressing production systems, such as recombinant subunit vaccines using an optimized A/Hanoi/30408/2005 H5N1 HA antigen (PAV3-HA) baculovirus (20), plasmid DNA vaccines (21), virus-like particle elicited robust humoral and/or cellular immune responses in vac- vaccines (22), and replication-incompetent rAd vectors (10,23,24). cinated mice; thus, a PAV3-based vector is capable of mediating In this study we generated an rAd that expressed the HA gene of swift, strong immune responses and offers a promising alternative H5N1 SIV using the Ad-Max adenovirus system. To overcome the to AdHu5 (31). The substantial progress in molecular engineering difficulties in screening for the rAd, we ligated the HA gene with of rAd vectors has indicated that AdHu26 as a vaccine vector may the EGFP gene via an IRES and then inserted the HA-IRES-EGFP suffer from limitations similar to those found for vectors based on fragment into an adenovirus vector, ensuring independent expres- other prevalent human adenoviruses (32). sion of nonrelated genes connected by an IRES (25). The rAd was In summary, rAd-H5HA-EGFP induced high levels of HI antibod- confirmed by PCR, RT-PCR, and Western blot analysis. The HA gene ies and provided protection against challenge with 2 different H5N1

124 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

influenza viruses. It is therefore a promising vaccine candidate for 13. Pacheco JM, Brum MC, Moraes MP, et al. Rapid protection of protection against animal influenza. Further studies now need to cattle from direct challenge with foot-and-mouth disease virus be conducted to evaluate the immune efficacy of the vaccine in (FMDV) by a single inoculation with an adenovirus-vectored pigs. FMDV subunit vaccine. Virology 2005;337:205–209. 14. Reed LJ, Muench H. A simple method of estimating fifty percent Acknowledgments endpoints. Am J Hyg 1938;27:493–497. 15. Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A This study was supported by the Harbin Municipal S&T Plan Laboratory Manual. 2nd ed. [Translated by Jing D and Li M]. (2009AA6BN078) and the Scientific Research Program of the State Beijing: Chinese Science Press, 1989:891–897. Key Laboratory of Veterinary Biotechnology (NKLVBP200818), 16. Chen H, Deng G, Li Z, et al. The evolution of H5N1 influenza Harbin, People’s Republic of China. viruses in ducks in southern China. Proc Natl Acad Sci U S A We thank Dr. Yanbing Li and Dr. Jianzhong Shi, Harbin Veterinary 2004;101:10452–10457. Research Institute, Chinese Academy of Agricultural Sciences, 17. World Organisation for Animal Health (Organisation mon- Harbin, for preparing the virus used in the animal challenge study. diale de la santé animale [OIE]). Manual of Diagnostic Tests and Vaccines for Terrestrial Animals. Paris, France: OIE, 2013. References Chapter 2.8.8, Swine influenza:6. 18. Ito T, Couceiro JN, Kelm S, et al. Molecular basis for the gen- 1. Fouchier RA, Munster V, Wallensten A, et al. Characterization of eration in pigs of influenza A viruses with pandemic potential. a novel influenza A virus hemagglutinin subtype (H16) obtained J Virol 1998;72:7367–7373. from black-headed gulls. J Virol 2005;79:2814–2822. 19. Holman DH, Wang D, Raja NU, et al. Multi-antigen vaccines 2. Kuiken T, Rimmelzwaan G, van Riel D, et al. Avian H5N1 influ- based on complex adenovirus vectors induce protective immune enza in cats. Science 2004;306:241. responses against H5N1 avian influenza viruses. Vaccine 3. Mushtaq MH, Juan H, Jiang P, et al. Complete genome analysis 2008;6:2627–2639. of a highly pathogenic H5N1 influenza A virus isolated from a 20. Treanor JJ, Wilkinson BE, Masseoud F, et al. Safety and immuno­ tiger in China. Arch Virol 2008;153:1569–1574. genicity of a recombinant hemagglutinin vaccine for H5 influ- 4. World Health Organization (L’Organisation mondiale de la enza in humans. Vaccine 2001;19:1732–1737. santé,[WHO]). Cumulative number of confirmed human cases 21. Kodihalli S, Goto H, Kobasa DL, et al. DNA vaccine encoding of avian influenza A(H5N1) reported to WHO,2003–2013. http:// hemagglutinin provides protective immunity against H5N1 www.who.int/entity/influenza/human_animal_interface/ influenza virus infection in mice. J Virol 1999;73:2094–2098. EN_GIP_20130215CumulativeNumberH5N1cases.pdf, WHO, 22. Perrone LA, Ahmad A, Veguilla V, et al. Intranasal vaccination 2013. with 1918 influenza virus-like particles protects mice and ferrets 5. Choi YK, Nguyen TD, Ozaki H, et al. Studies of H5N1 influenza from lethal 1918 and H5N1 influenza virus challenge. J Virol virus infection of pigs by using viruses isolated in Vietnam and 2009;83:5726–5734. Thailand in 2004. J Virol 2004;79:10821–10825. 23. Gao WT, Soloff AC, Lu X, et al. Protection of mice and poultry 6. Lipatov AS, Kwon YK, Sarmento LV, et al. Domestic pigs have from lethal H5N1 avian influenza virus through adenovirus- low susceptibility to H5N1 highly pathogenic avian influenza based immunization. J Virol 2006;80:1959–1964. viruses. PLoS Pathog 2008;4(7):e1000102. 24. Roy S, Kobinger GP, Lin J, et al. Partial protection against H5N1 7. Li H, Yu K, Yang H, et al. [Isolation and identification of H5N1 influenza in mice with a single dose of a chimpanzee adenovirus and H9N2 subtypes of avian influenza virus from Chinese pigs]. vector expressing nucleoprotein. Vaccine 2007;25:6845–6851. Chin J Prev Vet Med 2004;26:1–6. 25. Davies MV, Kaufman RJ. The sequence context of the initiation 8. Zhu Q, Yang H, Chen W, et al. A naturally occurring deletion codon in the encephalomyocarditis virus leader modulates effi- in its NS gene contributes to the attenuation of an H5N1 swine ciency of internal translation initiation. J Virol 1992;66:1924–1932. influenza virus in chickens. J Virol 2008;82:220–228. 26. Gubareva LV, McCullers JA, Bethell RC, et al. Characterization 9. Shi WF, Gibbs MJ, Zhang Y, et al. Genetic analysis of four porcine of influenza A/HongKong/156/97 (H5N1) virus in a mouse avian influenza viruses isolated from Shandong, China. Arch model and protective effect of zanamivir on H5N1 infection in Virol 2008;153:211–217. mice. J Infect Dis 1998;178:1592–1596. 10. Hoelscher MA, Garg S, Bangari DS, et al. Development of 27. McDermott JE, Shankaran H, Eisfeld AJ, et al. Conserved host adenoviral-vector-based pandemic influenza vaccine against response to highly pathogenic avian influenza virus infection in antigenically distinct human H5N1 strains in mice. Lancet 2006; human cell culture, mouse and macaque model systems. BMC 367:475–481. Syst Biol 2011;5:190. 11. Tatsis N, Ertl HC. Adenoviruses as vaccine vectors. Mol Ther 28. Uraki R, Kiso M, Shinya K, et al. Virulence determinants of pan- 2004;10:616–629. demic A(H1N1)2009 influenza virus in a mouse model. J Virol 12. Juillard V, Villefroy P, Godfrin D, et al. Long-term humoral 2013;87:2226–2233. and cellular immunity induced by a single immunization with 29. Tompkins SM, Zhao ZS, Lo CY, et al. Matrix protein 2 vaccina- replication-defective adenovirus recombinant vector. Eur J tion and protection against influenza viruses, including subtype Immunol 1995;25:3467–3473. H5N1. Emerg Infect Dis 2007;13:426–435.

2000;64:0–00 The Canadian Journal of Veterinary Research 125 FOR PERSONAL USE ONLY

30. Zhou D, Wu TL, Emmer KL, et al. Hexon-modified recombinant 32. Chen H, Xiang ZQ, Li Y, et al. Adenovirus-based vaccines: E1-deleted adenovirus vectors as dual specificity vaccine carriers Comparison of vectors from three species of adenoviridae. for influenza virus. Mol Ther 2013;21:696–706. J Virol 2010;84:10522–10532. 31. Patel A, Tikoo S, Kobinger G. A porcine adenovirus with low human seroprevalence is a promising alternative vaccine vector to human adenovirus 5 in an H5N1 virus disease model. PLoS One 2010;5(12):e15301.

126 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY Article

Porcine salivary analysis by 2-dimensional gel electrophoresis in 3 models of acute stress: A pilot study María Fuentes-Rubio, José J. Cerón, Carlos de Torre, Damián Escribano, Ana M. Gutiérrez, Fernando Tecles

Abstract The purpose of this research was to study changes in the salivary proteome of healthy pigs in stressful situations to identify any potential new salivary biomarker of stress. Three groups of animals were subjected to 3 stress models: snaring restraint followed by simulated sampling of vena cava blood; brief transport by road; and restriction of movement in a digestibility cage. Saliva was obtained from each animal before and 15 and 30 min after the induction of stress. The samples from the animals that showed the greatest increase in salivary cortisol concentration were pooled and run on 2-dimensional gels. Coomassie Brilliant Blue R-250 was used for spot detection and mass spectrometry for spot identification. Statistical analyses showed that 2 proteins had significant differences in expression before and after the induction of stress. These proteins were identified as odorant-binding protein and fragments of albumin. Further studies will be necessary to confirm the value of using these proteins as salivary biomarkers of stress in pigs.

Résumé L’objectif de la présente recherche était d’étudier les changements dans le protéome salivaire de porcs en santé dans des situations de stress afin d’identifier de nouveaux biomarqueurs de stress potentiels. Trois groupes d’animaux ont été soumis à 3 modèles de stress : contention au moyen d’un lasso suivie par simulation d’une ponction sanguine de la veine cave; bref transport sur route; et restriction des mouvements dans une cage à digestibilité. De la salive fut obtenue de chaque animal avant, ainsi que 15 et 30 minutes suivant l’induction du stress. Les échantillons provenant des animaux qui présentaient la plus grande augmentation de concentration de cortisol salivaire ont été regroupés et analysés sur gels en 2-dimensions. Le Bleu Brillant de Coomassie R-250 fut utilisé pour détection de taches et la spectrométrie de masse pour identification des taches. Les analyses statistiques ont montré que deux protéines avaient des différences significatives dans leur expression avant et après l’induction du stress. Ces protéines ont été identifiées comme étant une protéine de transport des odorants et des fragments de l’albumine. Des études ultérieures seront nécessaires pour confirmer la valeur d’utiliser ces protéines à titre de biomarqueurs salivaires du stress chez les porcs. (Traduit par Docteur Serge Messier) Introduction Several studies have shown that saliva samples can be used for the measurement of various biologic analytes that are useful for stress Stress is one of the most important factors to consider in animal monitoring, such as cortisol (7), chromogranin A (8), a-amylase (9), production for both animal-welfare and economic reasons. High and immunoglobulin A (10). Currently, cortisol is the most common levels of stress and poor welfare increase susceptibility to diseases, indicator of stress measured in farm animals, but its use as the only decrease life expectancy, impair growth and reproduction, damage stress indicator can have some limitations (11,12). For example, the the body, cause behavioral abnormalities (1), and decrease meat cortisol concentration depends not only on physical or psychological quality (2,3). stress but also on environmental conditions and metabolic factors Biomarkers of stress are commonly sought in blood or serum. (12). Therefore, it would be of great value to find new and sensitive Proteomic studies of porcine serum have shown that some proteins, biomarkers in saliva that could be useful to evaluate stress and such as haptoglobin, annexin IV, apolipoprotein A-I, and fetuin, welfare in the production chain. are biomarkers of welfare in these animals (4). In addition, other Proteomics has been used frequently in recent years to study proteins, such as serum amyloid A, paraoxonase 1, and platelet- biomarkers (13), and the results with saliva samples in pigs have activating factor acetylhydrolase, may act as markers of an oxidative been promising. A 2-dimensional gel electrophoresis (2-DE) map of stress response in pigs, as has been postulated in other animal spe- porcine saliva has recently been published (14) and used to identify cies (5). But blood sampling generates a substantial stress response biomarkers of disease in growing pigs under field conditions (15). by itself in pigs (6). For this reason, it is advisable to use samples The main objective of this research was to analyze possible whose collection is noninvasive and stress-free, such as saliva. changes in the salivary proteome of pigs suffering acute stress. The

Department of Animal Medicine and Surgery, Regional Campus of International Excellence Campus Mare Nostrum, University of Murcia, 30100, Espinardo, Murcia, Spain (Fuentes-Rubio, Cerón, Escribano, Gutiérrez, Tecles); Proteomics Unit, University Hospital Virgen de la Arrixaca, Ctra Madrid-Cartagena s/n 30120, El Palmar-Murcia, Spain (de Torre). Address all correspondence to Dr. Fernando Tecles; telephone: 134-868887082; fax: 134-868884147; e-mail: [email protected] Received February 7, 2013. Accepted June 13, 2013.

2014;78:127–132 The Canadian Journal of Veterinary Research 127 FOR PERSONAL USE ONLY

Table I. Differential protein expression before and 15 min after (T0 and T15) the induction of 3 different types of stress in 3 groups of pigs as evaluated with 2-dimensional electrophoresis Accession number Protein and Change in % spot in Swiss-Prot Theoretical MW Sequence pig group volume,T0 ➞ T15 database (kDa)/pI coverage (%) Mascot score 7, odorant-binding protein P81245 17.708/4.27 53.50 414.54 I 2 86% II 2 42% III 2 25% 22, albumin P08835 69.692/6.08 45.47 1490.01 I 1 85% II 1 16% III 1 3% MW — molecular weight; kDa — kilodaltons; pI — isoelectric point. pigs were subjected to 3 experimental models of acute stress: snaring The samples were obtained by introducing a small sponge into the restraint followed by simulated sampling of vena cava blood; brief pig’s mouth for at least 1 min with the help of a metal rod, as previ- transport by road; and restriction of movement in a digestibility cage. ously described (9). The sponges were placed in collection devices The salivary proteome profiles before and after the experimental (Salivette; Sarstedt, Aktiengesellschaft and Company, Nümbrecht, procedures were compared in a search for potential new biomarkers Germany) and centrifuged at 4000 3 g for 8 min to obtain the saliva. of stress. Data from this research could provide new tools to increase After centrifugation the saliva samples were stored at 280°C until the knowledge about changes in the expression of salivary proteins analyzed. in pigs under stress. All procedures involving the animals were approved by the University of Murcia Ethics Committee and followed the recommen- Materials and methods dations of the European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes (16). Animals and housing Salivary cortisol quantification Twenty-eight crossbred Duroc 3 [Landrace 3 Large White] grow- The salivary cortisol concentration was measured by means of the ing pigs (Sus scrofa domesticus) 3 to 4 mo of age from the experimental commercial kit Immulite 1000 Cortisol (Siemens Medical Solutions farm unit of the University of Murcia, Murcia, Spain, were used Diagnostics, Los Angeles, California, USA), validated for determina- in the study. The pigs were given access ad libitum to a nutrition- tions in porcine saliva (7). An increase in concentration was inter- ally balanced diet and water. They were housed in pens with a preted as a positive stress response (17). The Friedman test was done minimum space of 0.65 m2 per animal (16) and an average tempera- to assess the time point at which an increase in concentration in the ture of 23°C. The 3 experimental groups were housed in different stressed animals was statistically significant. All statistical analyses rooms of the farm and each animal was housed in a different pen were done with a statistical package GraphPad Prism, version 5.0 to avoid the influence of applied psychological stress on the other (GraphPad Software, La Jolla, California, USA). The significance animals. level was set at P , 0.05. Experimental procedures Proteomic analysis The animals were divided into 3 groups. The 10 pigs in group I In each group the samples from the 4 pigs with the highest sali- were subjected to snaring restraint followed by simulated sampling vary cortisol values were selected for proteomic evaluation. The time of vena cava blood; a nose sling was used for snaring and the lasso points selected for evaluation were T0 and T15, the latter correspond- kept under tension manually for 5 min. The 10 pigs in group II were ing to the time at which the highest concentrations were obtained. subjected to brief transport by road (40 min; 35 km) from the farm To reduce the influence of biologic variation the saliva specimens to the slaughterhouse. They were loaded into the trailer (which at T0 and T15 from the selected pigs in each group were pooled, for allowed 1.08 m2 of space per pig) without food or drink and were 2 pools per group. mixed to break social hierarchy. No stops were made along the way. Before the pooling, the salivary protein concentration of each The 8 pigs in group III were placed in a digestibility cage 1 m above sample was determined by the RC DC (reducing-agent and detergent- ground for 24 h; their space was sufficient only for turning in place, compatible) Protein Assay (BioRad, Hercules, California, USA). Before getting up, and lying down. electrophoresis, 150 mg of protein from the sample was treated with In all groups 3 saliva samples were taken from each animal; the the 2-D Clean-Up Kit (GE Healthcare Europe, Freiburg, Germany) and 1st sample was taken before the induction of stress (at T0), and the resuspended in rehydration buffer (30 mmol/L Tris basic, 2 mol/L remaining samples were taken 15 and 30 min later (at T15 and T30). thiourea, 7 mol/L urea, and 4% wt/vol CHAPS). The salivary protein

128 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

0.8 4 4

0.6 3 3 g/dL) g/dL) g/dL) m m 0.4 2 2

0.2 rtisol( 1 rtisol( 1 o C Cortisol( m Co 0.0 0 0 T0 T15 T30 T0 T15 T30 T0 T15 T30 Time point Time point Time point

Figure 1. Salivary cortisol concentrations in the 4 pigs in each group that showed the greatest response to 3 stressful situations: A — snaring restraint followed by simulated sampling of vena cava blood (group I); B — brief transport by road (group II); and C — restriction of movement in a digestibility cage (group III). Saliva samples were obtained before the induction of stress (at T0) and 15 and 30 min later (at T15 and T30). Each plot shows median (line within box), 25th and 75th percentiles (box), and 5th and 95th percentiles (whiskers). Each asterisk indicates a statistically significant difference (*P , 0.05) from T0. concentration was then determined again with the 2-D Quant Kit the protocol described by Shevchenko et al (20). The tryptic peptides (GE Healthcare Europe). Each sample pool contained 50 mg of protein were analyzed by capillary reversed-phase liquid chromatography from each of the 4 pigs, for a total of 200 mg per pool. coupled online with MS/MS. The column, BioBasic 18 (particle Each 2-DE gel was run in duplicate. Each pool was diluted in rehy- size 5 mm, pore size 300 Å, internal diameter 0.18 mm, and length dration solution (7 M urea, 2 M thiourea, 4% 3-[(3-­cholamidopropyl) 30 mm; Thermo Fisher Scientific, San Jose, California, USA), was dimethylammonio]-1-propanesulfonate [CHAPS], 60 mM dithioth- connected to a Surveyor MS Pump Plus (Thermo Fisher Scientific) reitol [DTT], 0.5% ampholytes [pH 3 to 10], and 0.002% bromophenol and then coupled with an LXQ Linear Ion Trap Mass Spectrometer blue) and loaded onto an 18-cm-long immobilized pH gradient strip (Thermo Fisher Scientific). The flow rate was set at 100 mL/min but (IPG), pH 3 to 10, during the 20-h rehydration period, according split to a rate of 2 mL/min. Mobile phase A was 0.1% formic acid/2% to the instructions of the manufacturer (GE Healthcare Europe). methanol in water and B was 0.1% formic acid in methanol. The Proteins were focused isoelectrically in the 1st dimension for a total peptide samples were injected and gradient elution was done under of 60 000 Vh in a horizontal electrophoresis chamber (Ettan IPGphor the following conditions: 5% B in 2 min; a linear increase of 5% to 3 system; GE Healthcare Europe). 70% B in 18 min; 70% B for 10 min; and 5% B for 15 min. The mass For the 2nd dimension the IPG strips were equilibrated for 15 min spectrometer was operated in a data-dependent MS/MS mode in with 2% DTT equilibration solution [6 M urea, 30% glycerol, 2% which the 15 most abundant peptide molecular ions in every scan sodium dodecyl sulfate (SDS), and 0.05 M Tris-HCl (pH 6.8)] and were sequentially selected for collision-induced dissociation with a then 15 min with 2.5% iodoacetamide (IAA) in equilibrate solution. normalized collision energy of 34%. Dynamic exclusion was applied Afterwards, SDS-polyacrylamide gel electrophoresis was done on to minimize repeated selection of peptides previously selected. The homemade 12% polyacrylamide gels measuring 170 3 170 3 1 mm capillary temperature and electrospray voltage were set to 200°C in a vertical electrophoresis chamber (Ettan DALTsix Electrophoresis and 3.5 kV. System; GE Healthcare Europe) at 5 mA per gel for 2 h and 2 W per The resulting mass spectra were searched against the Swiss-Prot gel for 7 h at 23°C (18). protein sequence database (453 320 sequences; released December 11, After electrophoresis the gels were fixed with a solution of 50% 2012) with Proteome Discoverer software, version 1.3 (Thermo Fisher (v/v) ethanol and 10% (v/v) acetic acid and stained with Coomassie Scientific), and the following search parameters: default charge states Brilliant Blue R-250 (19). Then 2-DE images were captured with the of 21, 31, and 41; a maximum of 1 missed cleavage; and an average ImageScanner III System (GE Healthcare Europe) and analyzed peptide mass tolerance of 1.5 Da. A fragment-ion search tolerance of by means of specific software (ImageMaster 2D Platinum 7.0; 0.8 Da was permitted. A fixed modification was carbamidomethyl- GE Healthcare Europe). Analysis included spot detection, landmark- ation of cysteine and a variable modification oxidation of methionine. ing, and spot matching of protein patterns for all gels in the set. All data were searched against a decoy database. The peptide score Percentage spot volume was analyzed for a difference between the cut-off for each run was automatically adjusted to ensure a 1% false- pigs in protein expression at T0 and T15 by a Wilcoxon signed-rank discovery rate throughout the experiments. A positive identification test and SPSS statistical software, version 10.0 (SPSS Inc., Chicago, was assigned when 2 or more unique peptides were identified. Illinois, USA). Spots with a P-value , 0.05 were manually picked from the stained gels for identification by mass spectrometry (MS). Results Analysis by mass spectronomy The salivary cortisol values for the 4 animals in each group with Proteins within the chosen gel spots were first reduced and alkyl- the greatest response to the stressor stimulus are shown in Figure 1. ated with DTT and IAA, respectively, and then digested to peptides The median values at T15 were significantly higher (P , 0.05) than with trypsin (Sigma-Aldrich, St. Louis, Missouri, USA) according to those at T0 in all the experimental groups: in group I those values

2000;64:0–00 The Canadian Journal of Veterinary Research 129 FOR PERSONAL USE ONLY

MW MW MW (kDa) (kDa) (kDa)

MW MW MW (kDa) (kDa) (kDa)

3 pI 10 3 pI 10 3 pI 10

Figure 2. Representative protein patterns obtained by 2-dimensional electrophoresis of saliva samples from experimental groups I, II, and III at T0 (A) and T15 (B). Significant spots identified by mass spectrometry are circled; the numbering corresponds with that in Table I. MW — molecular weight; kDa — kilodaltons; pI — isoelectric point. were 0.51 and 0.11 mg/dL, in group II 1.67 and 0.36 mg/dL, and in ing number of proteins will be investigated and discovered, with group III 0.92 and 0.30 mg/dL, respectively. the goal of maintaining and improving livestock productivity and The 2-DE gels obtained (Figure 2) were quite homogeneous despite welfare (22). A recent publication from proteomics research sug- being created from pools of samples obtained after different types of gested new porcine salivary biomarkers of disease conditions (15). stress, which could indicate that the pattern is quite stable despite However, to our knowledge no proteomic studies of the saliva of the many factors that can affect the composition of saliva. The gels pigs under stress had been reported before our study. were manually aligned by defining 4 landmarks well distributed over The choice of the stressful stimuli for this study was based on each entire image that clearly represented the same protein form. existing literature. Restraint with a nose sling, a technique for tempo- After statistical analysis of the percentage spot volume of each of the rarily restraining pigs that is commonly used in veterinary practice, 58 spots detected in all the gels, only 2 spots (nos. 7 and 22) appeared represents an acute stressor (23,24). Similarly, poor environmental to be differentially regulated by the experimental stressor at T15 in conditions during road transport are noted in the literature to be comparison with T0; the same behavior of these 2 spots was observed very stressful to animals (25,26). Digestibility cages are normally with all 3 stress-induction methods (Table I). Spot 7 decreased more used in studies of metabolism and nutritional evaluation (27,28); the than 1.5-fold with stress, the greatest decrease being in group I. In restriction of movement causes stress to the animal. contrast, spot 22 increased, the most pronounced change again being For proteomics purposes, the saliva samples from each group and in group I; however, in general the observed changes in this spot were time point were pooled. It has been postulated that the use of pools less than those observed in spot 7. With MS, spot 7 was identified as in proteomics studies could result in some loss of information in com- odorant-binding protein (OBP) and spot 22 as albumin. parison with the use of single samples (29). However, pooling is a valid and potentially valuable method to detect a common pattern of protein Discussion expression in the analysis of limited amounts of sample (30–32) and to eliminate the biologic variation between individuals (33). In addition, Saliva is a fluid with many research and diagnostic possibili- the total protein concentration of porcine saliva is very low, 70 times ties, and its composition of thousands of proteins and peptides of less than the concentration in other biologic samples, such as serum. glandular or blood origin is suggested to hold physiological and Thus, in our experience, large-volume samples are required to reach pathological information accessible by proteomic research (21). The appropriate concentrations for proteomic analysis. Therefore, the use state-of-the-art of salivary proteomics is in evolution, and a grow- of individual samples of low amounts of saliva can be a limitation.

130 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

Differential expression of proteins in saliva samples from pigs sub- References jected to stress was observed in 2 gel spots. One spot was identified as OBP; this protein appeared to be down-regulated in all 3 groups 1. Bonizzi L, Roncada P. Welfare and immune response. Vet Res after stress was applied. Other researchers have found a decreased Commun 2007;31:97–102. in OBP after the application of a stressor in bovine bronchoalveolar 2. Chaloupková H, Illmann G, Neuhauserová K, et al. Preweaning lavage fluid (34) and in porcine saliva from diseased animals (15). housing effects on behavior and physiological measures in pigs The OBPs, low-molecular-weight soluble proteins, are a subclass during the suckling and fattening periods. J Anim Sci 2007;85: of the lipocalin family (35,36). They are highly concentrated in the 1741–1749. nasal mucus of vertebrates (37), are excreted in biologic fluids, such 3. Guzik AC, Matthews JO, Kerr BJ, et al. Dietary tryptophan effects as saliva (38), and are thought to be involved in the release and on plasma and salivary cortisol and meat quality in pigs. J Anim modulation of chemical messages with pheromonal activity through Sci 2006;84:2251–2259. a still-undefined mechanism (37,39). 4. Roncada P, Begni B, Amadori M, et al. Blood serum proteome for Several hypotheses about why the OBP concentration decreases in welfare evaluation in pigs. Vet Res Commun 2007;31:321–325. stressful conditions could be made. Stress is known to be a factor in 5. Turk R, Piras C, Kovacic M, et al. Proteomics of inflammatory decreased reproductive activity and therefore could also lower the and oxidative stress response in cows with subclinical and clini- values of related proteins such as OBP. However, the mechanisms cal mastitis. J Proteomics 2012;75:4412–4428. by which stress impacts reproduction and OBP concentration are 6. Merlot E, Mounier AM, Prunier A. Endocrine response of gilts not clearly understood (40). Also, the increase in oxidative prod- to various common stressors: A comparison of indicators and ucts associated with stress could contribute to the decrease in OBP methods of analysis. Physiol Behav 2011;102:259–265. concentration (33). 7. Escribano D, Fuentes-Rubio M, Cerón JJ. Validation of an auto- Currently there are several commercial enzyme-linked immu- mated chemiluminescent immunoassay for salivary cortisol nosorbent assay kits for the detection of OBP in serum, plasma, measurements in pigs. J Vet Diagn Invest 2012;24:918–923. and biologic fluids from humans. However, whether there is cross- 8. Escribano D, Soler L, Gutiérrez AM, Martínez-Subiela S, Cerón JJ. reactivity with porcine OBP and whether these methods are sensitive Measurement of chromogranin A in porcine saliva: Validation enough to detect the concentrations of OBP in saliva have not been of a time-resolved immunofluorometric assay and evaluation confirmed. For this reason, it would be necessary to perform ana- of its application as a marker of acute stress. Animal 2012; lytical studies and clinical validation of these kits before they could 9:1–8. be used for porcine OBP quantification in saliva. If any of these kits 9. Fuentes M, Tecles F, Gutiérrez A, Otal J, Martínez-Subiela S, were analytically validated, they could be used for studies of stress Cerón JJ. Validation of an automated method for salivary alpha- involving large number of animals. amylase measurements in pigs and its application as a biomarker It is common that some proteins appear as multiple-spot chains of stress. J Vet Diagn Invest 2011;23:282–287. or fractions in different parts of the gel in salivary 2-DE gel images 10. Muneta Y, Yoshikawa T, Minagawa Y, Shibahara T, Maeda R, (15). In the present study, statistical analysis revealed changes in Omata Y. Salivary IgA as a useful non-invasive marker for spot 22, an albumin fragment. Other studies had similarly observed restraint stress in pigs. J Vet Med Sci 2010;72:1295–1300. differential expression of albumin fragments in situations of altered 11. Mormede P, Andanson S, Auperin B, et al. Exploration of the homeostasis, such as disease (15) or oxidative stress (41). Thus, hypothalamic–pituitary–adrenal function as a tool to evaluate some authors claim that albumin is a biomarker of the welfare of animal welfare. Physiol Behav 2007;92:317–339. pigs (4). It could be that high concentrations of cortisol can lead to 12. Jaskulke S, Manteuffel G. No apparent effect of an experimental an increase in albumin production, as has been reported for dogs narrow confinement on heart activity and cortisol in domestic (42). Therefore, it might be thought that in this study the high pigs. Animal 2011;5:433–438. levels of cortisol caused by induced stress increased the albumin 13. Gao J, Garulacan LA, Storm SM, et al. Biomarker discovery in levels. biological fluids. Methods 2005;35:291–302. This salivary proteomic study had some limitations, such as 14. Gutiérrez AM, Miller I, Hummel K, et al. Proteomic analysis of the use of pooled samples and the low number of animals; thus, it porcine saliva. Vet J 2011;187:356–362. should be considered preliminary. However, one can conclude from 15. Gutiérrez AM, Nöbauer K, Soler L, et al. Detection of potential the results that acute stress involves changes in saliva proteins such markers for systemic disease in saliva of pigs by proteomics: as OBP and fractions of albumin. Further studies involving more A pilot study. Vet Immunol Immunopathol 2013;151:73–82. animals and other stress conditions should be done to assess the 16. Council of Europe. European Convention for the Protection of value of these proteins as biomarkers for evaluating and monitoring Vertebrate Animals Used for Experimental and Other Scientific stress in pigs under field conditions. Purposes. ETS 123. Strasbourg, France: The Council, 1986. Available, with subsequent amendments, at http://conventions. Acknowledgments coe.int/treaty/en/treaties/html/123.htm Last accessed January 5, 2014. This study was supported by a PhD grant from the Spanish 17. Hellhammer DH, Wüst S, Kudielka BM. Salivary cortisol as Ministry of Education and by the Ministry of Economy and a biomarker in stress research. Psychoneuroendocrinology Competitiveness (AGL 2012-33612). 2009;34:163–171.

2000;64:0–00 The Canadian Journal of Veterinary Research 131 FOR PERSONAL USE ONLY

18. de Torre C, Ying SX, Munson PJ, Meduri GU, Suffredini AF. 31. Weinkauf M, Hiddemann W, Dreyling M. Sample pooling in Proteomic analysis of inflammatory biomarkers in bronchoal- 2-D gel electrophoresis: A new approach to reduce nonspecific veolar lavage. Proteomics 2006;6:3949–3957. expression background. Electrophoresis 2006;27:4555–4558. 19. Neuhoff V, Arold N, Taube D, Ehrhardt W. Improved staining 32. Neubauer H, Clare SE, Kurek R, et al. Breast cancer proteomics of proteins in polyacrylamide gels including isoelectric focus- by laser capture microdissection, sample pooling, 54-cm IPG IEF, ing gels with clear background at nanogram sensitivity using and differential iodine radioisotope detection. Electrophoresis Coomassie Brilliant Blue G-250 and R-250. Electrophoresis 2006;27:1840–1852. 1988;9:255–262. 33. Baldini C, Giusti L, Ciregia F, et al. Proteomic analysis of saliva: 20. Shevchenko A, Tomas H, Havlis J, Olsen JV, Mann M. In-gel A unique tool to distinguish primary Sjögren’s syndrome from digestion for mass spectrometric characterization of proteins secondary Sjögren’s syndrome and other sicca syndromes. and proteomes. Nat Protoc 2006;1:2856–2860. Arthritis Res Ther 2011;13:R194. Epub 2011 Nov 25. 21. Lamy E, Mau M. Saliva proteomics as an emerging, non-invasive 34. Mitchell GB, Clark ME, Siwicky M, Caswell JL. Stress alters the tool to study livestock physiology, nutrition and diseases. cellular and proteomic compartments of bovine bronchoalveolar J Proteomics 2012;75:4251–4258. lavage fluid. Vet Immunol Immunopathol 2008;125:111–125. 22. Scarano E, Flortita A, Picciotti PM, et al. Proteomics of saliva: 35. Staiano M, D’Auria S, Varriale A, et al. Stability and dynamics Personal experience. Acta Otorhinolaryngol Ital 2010;30:125–130. of the porcine odorant-binding protein. Biochemistry 2007;46: 23. Geverink NA, Schoutenb WGP, Gortc G, Wiegant VM. Individual 11120–11127. differences in behavioral and physiological responses to restraint 36. Pelosi P. Odorant-binding proteins: Structural aspects. Ann N stress in pigs. Physiol Behav 2002;77:451–457. York Acad Sci 1998;855:281–293. 24. Rushen J, Ladewig J. Stress-induced hypoalgesia and opioid 37. Pelosi P. Odorant-binding proteins. Crit Rev Biochem Mol Biol inhibition of pigs’ responses to restraint. Physiol Behav 1991; 1994;29:199–228. 50:1093–1096. 38. Marchese S, Pes D, Scaloni A, Carbone V, Pelosi P. Lipocalins 25. Bao E, Sultan KR, Nowak B, Hartung J. Expression and distribu- of boar salivary glands binding odours and pheromones. Eur J tion of heat shock proteins in the heart of transported pigs. Cell Biochem 1998;252:563–568. Stress Chaperones 2008;13:459–466. 39. Tegoni M, Pelosi P, Vincent F, et al. Mammalian odorant binding 26. Warriss PD, Brown SN, Adams SJM, Corlett IK. Relationship proteins. Biochim Biophys Acta 2000;1482:229–240. between subjective and objective assessments of stress at slaugh- 40. Tilbrook AJ, Turner AI, Clarke IJ. Effects of stress on reproduc- ter and meat quality in pigs. Meat Sci 1994;38:329–340. tion in non-rodent mammals: The role of glucocorticoids and 27. Sulabo RC, Stein HH. Digestibility of phosphorus and cal- sex differences. Rev Reprod 2000;5:105–113. cium in meat and bone meal fed to growing pigs. J Anim Sci 41. Marco-Ramell A, Arroyoa L, Sacoa Y, et al. Proteomic analysis 2013;91:1285–1294. Epub 2012 Dec 10. reveals oxidative stress response as the main adaptative physi- 28. Li SF, Niu YB, Liu JS, et al. Energy, amino acid, and phosphorus ological mechanism in cows under different production systems. digestibility of phytase transgenic corn for growing pigs. J Anim J Proteomics 2012;75:4399–4411. Sci 2013;91:298–308. Epub 2012 Oct 16. 42. Caldin M, Tasca S, Carlo E, et al. Serum acute phase protein 29. Zolg W. The proteomic search for diagnostic biomarkers. Lost concentrations in dogs with hyperadrenocorticism with and in translation? Mol Cell Proteomics 2006;5:1720–1726. without concurrent inflammatory conditions. Vet Clin Pathol 30. Diz AP, Truebano M, Skibinski DO. The consequences of sample 2009;38:63–68. pooling in proteomics: An empirical study. Electrophoresis 2009; 30:2967–2975.

132 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY Article

Specific contrast ultrasound using sterically stabilized microbubbles for early diagnosis of thromboembolic disease in a rabbit model Michal Vlašín, Robert Lukáˇc, Zuzana Kauerová, Pavel Kohout, Josef Mašek, Eliška Bartheldyová, Št ˇepán Koudelka, Zina Korvasová, Jana Plocková, Nikola Hronová, Jaroslav Turánek

Abstract Specific contrast ultrasound is widely applied in diagnostic procedures on humans but remains underused in veterinary medicine. The objective of this study was to evaluate the use of microbubble-based contrast for rapid ultrasonographic diagnosis of thrombosis in small animals, using male New Zealand white rabbits (average weight about 3.5 kg) as a model. It was hypothesized that the use of microbubble-based contrast agents will result in a faster and more precise diagnosis in our model of thrombosis. A pro-coagulant environment had been previously established by combining endothelial denudation and external vessel wall damage. Visualization of thrombi was achieved by application of contrast microbubbles [sterically stabilized, phospholipid-based microbubbles filled with sulfur hexafluoride (SF6) gas] and ultrasonography. As a result, rapid and clear diagnosis of thrombi in aorta abdominalis was achieved within 10 to 30 s (mean: 17.3 s) by applying microbubbles as an ultrasound contrast medium. In the control group, diagnosis was not possible or took 90 to 180 s. Therefore, sterically stabilized microbubbles were found to be a suitable contrast agent for the rapid diagnosis of thrombi in an experimental model in rabbits. This contrast agent could be of practical importance in small animal practice for rapid diagnosis of thrombosis.

Résumé L’échographie par contraste spécifique est une procédure diagnostique couramment utilisée chez les humains mais demeure sous-utilisée chez les animaux. L’objectif de la présente étude était d’évaluer l’utilisation du contraste basée sur les micro-bulles pour le diagnostic échographique rapide de thrombose chez les petits animaux, en utilisant comme modèle le lapin blanc de Nouvelle-Zélande mâle (poids moyen de 3,5 kg). L’hypothèse a été émise que l’utilisation d’agents de contraste à base de micro-bulles résulterait en un diagnostic plus rapide et plus précis dans notre modèle de thrombose. Un environnement pro-coagulant a préalablement été établi en combinant le dénudement endothélial et du dommage à la paroi externe du vaisseau. La visualisation des thrombi a été obtenue par application de micro-bulles de contraste [micro-bulles

à base de phospholipides remplies d’hexafluorure de soufre (SF6) stabilisées stériquement] et échographie. L’application de micro-bulles comme milieu de contraste pour l’échographie résulta en un diagnostic rapide et clair de thrombi dans l’aorte abdominale en 10 à 30 secondes (moyenne de 17,3 s). Dans le groupe témoin, le diagnostic n’était pas possible ou prenait de 90 à 180 s. Ainsi, des micro-bulles stabilisées stériquement ont été trouvées comme étant un agent de contraste convenable pour le diagnostic rapide de thrombi dans un modèle expérimental chez les lapins. Cet agent de contraste pourrait être d’importance concrète en pratique des petits animaux pour le diagnostic rapide de thromboses. (Traduit par Docteur Serge Messier)

Introduction tinal tract, and brain, can also be affected. Vasoactive agents, such as prostaglandin or serotonin, that are released by platelets at the site of Systemic thromboembolism is a frequent and life-threatening com- the thrombus result in constriction of collateral and regional vessels, plication of cardiomyopathy in cats. Stasis of blood within dilated which further contributes to ischemia and reduces the blood flow to cardiac chambers, turbulent flow, and increased platelet reactivity terminal spinal cord segments (1,2). A saddle thrombus results in a combine to predispose a cardiomyopathic patient to systemic throm- consistent constellation of physical abnormalities, including a pelvic boembolism (1). Typically, the clot lodges at the aortic furcation limb paresis or paralysis, absence of femoral pulsation, cyanosis, and (saddle thrombus), which results in a severe ischemic insult to the local hypothermia. Hind limb musculature is typically firm and pain- pelvic limbs and/or tail. A smaller clot may enter 1 iliac artery and ful. Organ dysfunction may occur depending on the location of the cause unilateral paresis or paralysis of the limb. Besides pelvic limbs, thrombus. Affected cats almost always have significant underlying 1 thoracic limb and certain organs, such as the kidneys, gastrointes- cardiac disease and sometimes even heart failure. Overall prognosis

International Clinical Research Center (ICRC), St. Anne’s University Hospital Brno, Pekaˇrská 53, 656 91 Brno, Czech Republic (Vlašín, Kohout, Hronová); Department of Pharmacology, and Immunotherapy, Veterinary Research Institute, Hudcova 70, 621 00 Brno, Czech Republic (Lukáˇc, Kauerová, Mašek, Bartheldyová, Koudelka, Korvasová, Plocková, Turánek); Section of Small Animal Diseases, University of Veterinary and Pharmaceutical Sciences, Palackého 1-3, 612 42 Brno, Czech Republic (Vlašín, Kohout, Hronová). Address all correspondence to Dr. Jaroslav Turánek; telephone: (420) 5-3333-1311; fax: 1420 5 4121 1229; e-mail: [email protected] Received May 14, 2012. Accepted March 25, 2013.

2014;78:133–139 The Canadian Journal of Veterinary Research 133 FOR PERSONAL USE ONLY

of this complication is guarded. Approximately 50% of affected cats Alabaster, Alabama, USA) consisting of 1,2-dipalmitoyl-sn- do not survive the congestive heart failure and systemic thrombo- glycero-3-phosphocholine (DPPC), 1,2-distearoyl-sn-glycero-3-­ embolic crisis and die within 6 to 36 h. Those surviving typically phosphoethanolamine-N-[poly(ethyleneglycol)2000] (PEG2000DSPE), show steady improvement in limb function within 24 to 72 h after and Tween 80 (Sigma-Aldrich, St. Louis, Missouri, USA). Lipids were presentation. The prognosis is poor for cats not showing an early dissolved in chloroform and mixed properly. The mixture was sub- improvement and euthanasia should be considered (1–4). sequently transferred to a round-bottom flask and the solvent was Regardless of the underlying cardiac disease, local control of removed by rotary evaporation at 37°C (Laborota 4000, Heidolph, the thrombus must be achieved within a relatively short period of Germany). The lipid film was hydrated by adding phosphate buffer time. Symptoms usually begin with pelvic paresis or paralysis with saline (PBS) solution up to the final lipid concentration of 3 mg/mL marked pain. Absence of femoral pulsation is a typical sign but to above the transition temperature of the used lipids. Afterwards, make a decision, an early and proper diagnosis is essential (1,2). In DPPC liposomes were rapidly frozen in liquid nitrogen and thawed general, small-animal practice angiography is typically unavailable 5 times in a water bath at 55°C. The resulting liposome suspension and Doppler abdominal ultrasonography requires an experienced was extruded through 400-nm polycarbonate membrane filters at radiologist in order to confirm the arterial thrombosis beyond a 55°C using a Mini-Extruder (Avanti Polar Lipids). Thereafter, the size reasonable doubt (1–4). For better visualization of the distal aorta distribution of liposome suspension was determined by Zetasizer (dorsal to the urinary bladder), we suggest using novel contrast NanoZS (Malvern, Worchester, UK). media to make the thrombus more transparent. Although specific The liposome suspension (1 mL) was transferred to a hermetic contrast ultrasound is widely applied in diagnostic procedures on 1.5-mL vial. The vial was filled with sulfur hexafluoride (SF6) gas humans, it is still underused in veterinary medicine. Early and pre- (Messer, Gumpoldskirchen, Austria) and mixed intensively for cise diagnosis of thromboembolic disease in cats could potentially 30 s using the capsule mixing device 3M ESPE CapMix (3M ESPE, improve the treatment, thus resulting in a better prognosis. Feline Germany) (13). The final composition of the MB sample was patients that are treated earlier in the disease process might benefit 2 mg/mL DPPC, 1 mg/mL PEG2000DSPE, and 0.03 mg/mL Tween 80. from an improved quality of life. Ultrasound contrast agents are currently used to improve visual- Microbubble characterization by optical ization of the microvasculature within organs and vital structures. microscopy These substances usually consist of small gas microbubbles (MBs) The size and concentration of MBs were determined by optical stabilized by a surfactant (5,6). Surfactants used include serum microscopy (14). The MB samples were taken directly from the albumin, polymers, and phospholipids (7). Lipid-coated, gas-filled vial, diluted 50 times with PBS, and the final volume of 10 mL was MBs represent a new class of drug delivery system with both diag- transferred into the Bürker Counting Chamber. An Eclipse TE200 nostic and therapeutic application (8). Lipid-based carrier systems microscope (Nikon, Japan) was employed and the magnification represent drug vehicles composed of physiological lipids such as of the objective Nikon LWD 203 was used to observe and capture phospholipids, cholesterol, cholesterol esters, and triglycerides, as the images of MBs. The images were analyzed by LUCIA software well as synthetic auxiliary lipids that provide the lipid particles, (Laboratory Imaging, Czech Republic) to determine the absolute especially their surface, with a special function. The main advan- count and size distribution of MBs. tage of MB application as drug delivery systems is a reduction of undesired side effects such as toxicity owing to drug targeting (9). Static light scattering (SLS) Furthermore, diagnostic application of ultrasound imaging using HORIBA’s LA-950 Laser Diffraction Particle Size Distribution MBs has become very popular because ultrasound is a noninvasive Analyzer (Horiba, France) was used to determine the MB size distri- and relatively low-cost diagnostic tool. It uses portable, real-time bution by static light scattering (SLS) (14). At first, the method was imaging equipment and also avoids hazardous ionizing radiation optimized by Megabead NIST Traceable Particle Size Standards from (9,10). Microbubbles are small microspheres (typically 1 to 8 mm 1, 3, 6, 10, to 15 mm (Polysciences, Warrington, Pennsylvania, USA). in diameter) filled with high-molecular-weight gases such as per- The 1.0% suspension of polystyrene microspheres in water was fluorocarbons or sulfur hexafluoride (SF6) that result in decreased diluted by degassed and filtered PBS and measured. An MB sample solubility and prolonged lifespan of MBs within the circulation (9). (5 to 30 mL, 3 mg phospholipid/mL) was injected into a cuvette-type The objective of the present study was to evaluate microbubble- fraction cell (filled with 10 mL of PBS), equipped with a magnetic based contrast for rapid ultrasonographic diagnosis of thrombosis stirrer to prevent non-homogenous distribution owing to flotation of in small animals using male New Zealand white rabbits (11,12). We MBs. All samples were measured immediately after the application hypothesize that the use of microbubble-based contrast agents will into the cuvette and were analyzed both for number- and volume- result in a faster and more precise diagnosis in this model. weighted size distribution. Animals Materials and methods Male New Zealand white rabbits (approximately 6 mo old, average weight 3.5 kg) were purchased from Biotest Konarovice Preparation of liposomes and microbubbles (Konarovice, Czech Republic) and housed for 14 d before the Liposomes were prepared by the lipid film hydration method. experiments. The animals were treated in accordance with the In brief, liposomes were composed of lipids (Avanti Polar Lipids, “Guide for Care and Use of Laboratory Animals” (DHHS publ.

134 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

Table I. System for assessing clot visibility Clot transparency Description 0 No clot identified 1 Suspected clot with no detectable contours 11 Distinctive clot, yet still doubtful contours 111 Evident clot with ill-defined contours 1111 Evident clot with well-defined contours

Table II. Parameters of microbubbles (MB) preparation MB size (mm) by optical microscopy Average 6 SD 2.11 6 1.41 Min 0.66 Max 12.10 Figure 1. Arterial thrombosis rabbit model. Complete isolation of distal aorta with mural clot (arrow) that can be seen after 30 min of maturation. MB size (mm) by SLS Average 6 SD 3.52 6 0.68 Min 1.94 No. NIH 85-23, revised 1996, Office of Science and Health Reports, Max 10.25 Bethesda, Maryland, USA). The experimental protocol was approved MB concentration (MBs/mL) by the Ethical Committee at the University of Veterinary and Optical microscopy 2.33 3 109 Pharmaceutical Sciences in Brno. Cell counter 2.91 3 109 Animal groups SD — Standard deviation; SLS — Static light scattering. Composition of MBs was 2 mg/mL of DPPC 1 1 mg/mL of The rabbits were randomized into 2 groups (n = 6), as we intended PEG DSPE 1 0.03 mg/mL of Tween 80. to precisely establish the time up to clot identification. 2000

Anesthesia and monitoring of the abdominal aorta via the left iliac artery and kept in place by Anesthesia was induced with a mixture of 0.2 mg/kg body weight 2 tourniquets occluding the vascular segment, placed approximately (BW) of medetomidine and 15 mg/kg BW of ketamine mixed in 10 mL apart. A pro-coagulant environment had been previously 1 syringe and administered as a single intramuscular injection into established by combining endothelial denudation and external ves- the dorsal lumbar muscles. A temporary tracheostomy was done sel wall damage. The clot was left in place for 30 min of maturation using a number-4 tracheal tube without cuff and anesthesia was (Figure 1) and the tourniquets were then removed and an external maintained by a mixture of oxygen and isoflurane ranging from constrictor (over the needle loose 4-0 silk ligation) was applied 1% to 1.5% of isoflurane. A 22G catheter was inserted percutane- distally to prevent the clot from moving. After the tourniquet was ously into the marginal auricular vein and constant rate infusion of released, the flow was partially re-established. With the use of flow Ringer’s solution was administered using a perfusor at the rate of probes, the blood flow distal to the thrombus was monitored for 10 mL/kg per hour. The left common carotid artery was then cannu- 10 min. The flow probes were then removed and organs were flushed lated and the arterial line was connected to a pre-calibrated arterial using warm sterile saline solution. The abdominal wall was then blood pressure transducer for continuous recording of blood pres- closed in a simple continuous suture pattern (3-0 Ethibond, Ethicon) sure and heart rate. The circulatory and respiratory variables evalu- using a single layer. ated included the systolic (SAP), mean (MAP), and diastolic (DAP) Endothelial denudation was achieved by passing an inflated #4 arterial blood pressure, heart rate (HR), saturation of hemoglobin Fogarty embolectomy catheter 30 times through an isolated aortic by oxygen (SpO2), and respiratory rate (RR). Systolic arterial blood segment via cannulation of the left external iliac artery. Aortic ste- pressure (SAP), MAP, and DAP were measured directly through an nosis was then achieved by use of an external constrictor and the arterial blood pressure transducer, HR was measured by electrocar- lumen reduction was maintained at 295% (as assessed by angiogra- diogram (ECG) electrodes applied on the thorax, SpO2 by a sensor phy). External damage of the aortic wall was induced by 16 clamps applied on the tongue, and RR was measured electronically based on proximal to the stenosis using straight Mosquito hemostat. Blood thorax-impedance changes. All parameters were acquired and saved pressure and aortic blood flow were recorded by use of a Single by a Mindray PM-9000Vet vital monitor (Mindray, Shenzhen, China). Channel Flow Meter. Thereafter, distal aortic thrombus was identified by ultrasonog- Surgery raphy with Doppler system (Vivid 7; GE Healthcare, Pittsburgh, We modified a previously validated and published rabbit model Pennsylvania, USA). To evaluate the efficacy of the MB suspen- system (11,12) that simulated the conditions of acute arterial throm- sion as a contrast substance, 300 mL of saline was intravenously bosis. In brief, thrombosis was induced by injecting 10 mL of whole administered in 1 group, while MB suspension was applied to the blood obtained from the carotid arterial line into the distal segment venous system of the other group. The time to clot identification was

2000;64:0–00 The Canadian Journal of Veterinary Research 135 FOR PERSONAL USE ONLY

Figure 2. Characterization of microbubbles. A — Optical microscopy of microbubbles. Composition was 2 mg/mL DPPC, 1 mg/mL

PEG2000DSPE, and 0.03 mg/mL TWEEN 80. B — Transmission electron microscopy of microbubbles. During desiccation of the sample in the vacuum inside the electron microscope, the microbubbles burst and the ruptures could be seen in their membranes. C — Graph of the size distribution of microbubbles measured by static light scattering (Horiba LA 950 SLS instrument). D — Histogram of microbubbles obtained by optical microscopy.

­measured and the transparency of the image was assessed in each undertaken for explorative comparison in order to complete results case. The MBs were administered through the venous line in 1 bolus of descriptive statistics. P-values lower than 0.05 were considered at a dose of 300 mL (approximately 7 3 108 MBs, 0.9 mg of lipid). to be statistically significant. The study was blinded so the ultrasonographer was not aware of the treatment applied to the specific animal. Results After termination of each experiment, the animal was euthanized under anesthesia using a high dose of sodium pentobarbital, a suffi- cient aortic segment was explanted, and the mural clot was carefully Characterization of microbubbles removed from the lumen. It was weighed (mg) immediately and the The size distribution of MBs was determined by static light scat- clot size was compared with the ultrasonography (USG) image with tering (SLS) and by optical microscopy using the HORIBA-LA 950 and without the MBs. (HORIBA Scientific, Edison, New Jersey, USA). Its main advantage over other SLS instruments is that measurements can be done in a Endpoints relatively small cuvette (10 mL) with magnetic stirring to prevent As an index of efficacy, i) time to clot identification (s); ii) transpar- changes in the distribution of MBs in the laser beam area due to ency of the image (in the range of 1 to 1111); and iii) the weight flotation. The size distribution of the MBs (expressed in terms of of the mural clot (mg) were documented. In all the experiments, the number) ranged from approximately 2 to 10 mm with the mean clot identification was done by the same observer, an experienced 3.52 6 0.68 mm. This data correlates well with the optical microscopy clinician, who was blinded to the treatment groups. The system for data. The average size of MBs determined by optical microscopy assessing the clot visibility is outlined in Table I. was 2.11 6 1.41 mm, with the size ranging from 0.66 to 12.10 mm (Table II, Figure 2). Optical microscopy allowed prepared MBs to Statistical analysis be directly observed. The multimodal distribution was recognized Statistical analysis was carried out using statistical software by both techniques (Figure 2). Statistica 9.1 (StatSoft, Tulsa, Oklahoma, USA). The number of Optical microscopy and the cell counter BC-2800 VET (Mindray) observations in this study is on a lower margin where it is appropri- were used to evaluate the concentration of MBs in the prepared ate to use statistical tests. Nonparametric Mann-Whitney U-test is samples. The counting of MBs revealed that the amount of MBs was

136 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

Figure 3. Image of the clot visualized by microbubble contrast media. A — The ultrasound image of the abdominal vessel with stenosis before the application of microbubbles. B — The ultrasound image of the abdominal vessel with stenosis after the appli- cation of microbubbles. Long white arrow indicates the direction of blood flow; short black thick arrows indicate the vessel wall; long black arrows indicate stenosis; broken white arrow indicates direction of blood flow.

2.33 3 109 MBs/mL and 2.91 3 109 MBs/mL for optical microscopy clot identification in the study group suggests that the variability and the cell counter, respectively (Table II). of the studied method is quite low. Due to the very large difference between the groups, a small sample size was enough to demonstrate, Animal study significantly (P = 0.0027), the lower time needed to identify the clot The mean clot weight in a control group was 116.7 (6 16.33) mg when microbubbles are used. The weight of the clots was well bal- with the range of 90 to 130 mg. However, the clot was identified anced between the groups. The difference in median of the weights beyond a reasonable doubt in only 4 out of 6 animals with a transpar- was only 5 mg and the range was also very much alike. ency of 1 star in all detectable cases (Figure 3). Also, the time needed It is also evident from the obtained results that the transparency for clot identification varied from 90 to 180 s (mean: 148.8 6 41.31 s) of the image increases when microbubbles are applied. With the in the control group (Table III). microbubbles, two thirds of images are evaluated by the highest On the other hand, we succeeded in detecting each and every clot degree of transparency. In the control group; however, two thirds introduced to the study group. While the mean clot weight using of images are evaluated as the least transparent and the clot was not MB substance applied to the venous system was 118.3 (6 20.41) mg found at all in the remaining one third of images. with the range of 90 to 150 mg, the mean time to clot identifica- tion was significantly (P = 0.0027) shorter (10 to 30 s with mean of Discussion 17.3 6 7.15 s) than the control group. The transparency of the images also differed from the control, since it varied between 31 and 41 Some criteria need to be met in order to establish a new applicable (Table III). The ultrasonography was carried out in a standard man- diagnostic tool for an acute thromboembolia in small animals using ner and did not require any special manipulation with the probe. contrast ultrasonography. First, the procedure itself should not be The images were clear and easily recognizable, beyond suspicion technically demanding and the contrast media must be readily avail- (Figure 4). able, as thromboembolic disease is often an acute condition, with the patients arriving late in the afternoon or during the night, in an Statistical description of the results emergency situation. Lyophilized MB preparations sealed under the As shown in the statistics in Table III and in the values of the filling gas meet these criteria and are ready for rapid reconstitution observed parameters themselves, time to clot identification and and administration. Moreover, after reconstitution, they are stable transparency of the image differed between the study group and the for at least a working day when stored in a refrigerator under the control group. The application of microbubbles as an ultrasound con- filling gas. trast medium decreased the time to clot identification by 10 times. Second, intravenous (IV) administration of contrast media is The standard deviation and the variance in the time needed for preferred during the examination, since femoral artery cannulation

2000;64:0–00 The Canadian Journal of Veterinary Research 137 FOR PERSONAL USE ONLY

Table III. Parameters of thrombi identification in animal groups Study group Control group Number TCI (s) TI CW (mg) TCI (s) TI CW (mg) 1 15 1111 110 90 1 130 2 17 111 90 175 1 110 3 20 111 130 150 1 90 4 30 1111 150 N/A 0 130 5 10 1111 110 N/A 0 110 6 12 1111 120 180 1 130 N 6 6 4 6 Mean (SD) 17.3 (7.15) 118.3 (20.41) 148.8 (41.31) 116.7 (16.33) Median 16 115 162.5 120 Range 10 to 30 90 to 150 90 to 180 90 to 130 TCI — time to clot Identification (s); TI — (0 to 111) — transparency of the image; CW — weight of the mural clot at the end of each experiment (mg) for control (given 300 mL of saline) and study group (given 300 mL of contrast media) for clot imaging; N — number of animals; SD — standard deviation.

1 or more observers to assess the clot transparency would benefit the study. Also, while we are aware that larger groups of animals would most likely increase the statistical significance, we kept the numbers of animals as low as possible for ethical reasons. In our opinion, the differences between the groups are evident even when the lowest comparable numbers were used. Although selective angiography with iodinated contrast medium has been the gold standard for diagnosing aortic thrombosis, abdom- inal ultrasonography is the most common accessible method in identifying aortic thrombi (15). The ultrasonographic appearance of a thrombus depends on its duration. Chronic thrombi usually appear echogenic and heterogeneous, whereas immature thrombi Figure 4. Image of the clot visualized by microbubble contrast media. can be difficult to distinguish from flowing blood because they can Black arrow and black brackets indicate thrombus; white arrows indicate be hypoechoic to anechoic and homogeneous (16). Doppler flow the vessel walls of aorta abdominalis; white bracket indicates stenosis; evaluation is generally required to evaluate thrombi (15,17). broken white arrow indicates direction of blood flow. Lack of a Doppler signal is a diagnostic sign of a thrombus. In these cases, it is important that the settings of the ultrasound is not a routine procedure in a small animal practice, especially machine are correct and that optimal images of the affected vessels when the blood flow through the artery has been compromised. are obtained. Otherwise, the lack of a Doppler signal due to poor The contrast media must cross the pulmonary barrier through the technique could be misinterpreted as a sign of thrombosis (18). right cardiac outflow tract in order to enter the arterial circulation. When the conventional color or power Doppler techniques are not For clinical availability, IV administration has been recommended, sensitive enough, namely in cases of deeply located and/or small stressing the need for quick and easy passage through pulmonary vessels with slow flow or in vessels with inadequate Doppler angle, circulation as was proved in an earlier study in mice (13). ultrasound contrast media may be used to justify the presence or Third, the brightness of the image must be strongly superior to absence as well as the direction of blood flow in a certain vessel. any other method, showing both high sensitivity and specificity. We Although clinical application of commercially available contrast believe that this was the case in our study. agents in small animals may be beneficial in diagnosing aortic Even though the study was blinded, i.e., the ultrasonographer thromboembolism, its widespread use might be limited because of was not aware of the substance applied to the venous system, either the high price of the contrast media and the ultrasound instruments saline or MB suspension, the difference in image was apparent that contain contrast-specific second harmonic modalities (19). from the beginning. The time to clot identification was significantly To the authors’ knowledge, there are no reports of cases of aortic shorter and 100% of embolic clots introduced into the abdominal thromboembolism in small animals routinely diagnosed by the use aorta had been identified using our MB suspension. It took signifi- of (non-commercial) contrast agent and general ultrasonography cantly less time to obtain enough data to clearly diagnose the study without the use of Doppler. Therefore, a diagnosis based on applying group despite the fact that the difference in the mean explanted-clot a less expensive agent that provides a good contrast even with the weight was not statistically significant between the groups. Since we use of conventional ultrasound instrument modality might be very are looking for clinical relevance, we do not feel that the addition of promising due to its common accessibility.

138 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

In conclusion, acute thromboembolism is a potentially life-­ Optison, and Levovist in mice — Initial results. Radiology 2003; threatening condition that requires an immediate and efficient action. 229:423–428. Besides general diagnosis, the early and definitive identification of a 7. Pancholi KP, Farook U, Moaleji R, Stride E, Edirisinghe MJ. clot is absolutely crucial for future planning. In this study, we proved Novel methods for preparing phospholipid coated microbubbles. the successful application of microbubble-based contrast agents for Eur Biophys J 2008;37:515–520. rapidly diagnosing thrombosis in a rabbit model. Our plan was to 8. Rawat M, Singh D, Saraf S, Saraf S. Lipid carriers: A versatile develop and introduce a safe, easy-to-use, and efficient diagnostic delivery vehicle for proteins and peptides. Yakugaku Zasshi tool to help to localize clots as well as for possible use in therapeutic 2008;128:269−280. targeting. In our opinion, the MBs tested meet these criteria and, 9. Hernot S, Klibanov AL. Microbubbles in ultrasound-triggered among other possible uses, should be considered in the first-choice drug and gene delivery. Adv Drug Deliv Rev 2008;60:1153−1166. diagnostic plan for acute thromboembolic disease. 10. Liu Y, Miyoshi H, Nakamura M. Encapsulated ultrasound microbubbles: Therapeutic application in drug/gene delivery. Acknowledgments J Control Release 2006;114:89−99. 11. Chan AK, Rak J, Berry L, et al. Antithrombin-heparin covalent This study was supported by the following grants: Grant Agency of complex: A possible alternative to heparin for arterial thrombosis the Academy of Sciences of the Czech Republic GAAV KAN200520703 prevention. Circulation 2002;106:261−265. to J.T. and M.V., KAN200100801, and GAP503/12/G147 to J.T., 12. Klement P, Liao P, Bajzar L. A novel approach to arterial throm- Grant No. MZE 0002716202, the European Regional Development bolysis. Blood 1999;94:2735−2743. Fund FNUSA-ICRC No. CZ.1.05/1.1.00/02.0123 to M.V., and 13. Lukáˇc R, Kauerová Z, Mašek J, et al. Preparation of metallo- CZ.1.07/2.3.00/20.0164 to J.T. The authors thank Vladimir Babak chelating microbubbles and study on their site-specific interac- for statistical analysis and Pavla Filakova and Lucie Papschova for tion with rGFP-HisTag as a model protein. Langmuir 2011;27: technical assistance. 4829−4837. 14. Sennoga CA, Yeh JSM, Alter J, et al. Evaluation of methods for References sizing and counting of ultrasound contrast agents. Ultrasound Med Biol 2012;38:834−845. 1. Tilley LP. Thromboembolic disease. In: Northsworthy GD, ed. 15. Fox PR, Petrie JP, Hohenhaus AE. Peripheral vascular disease. The Feline Patient. Ames, Iowa: Blackwell Publishing, 2011: In: Ettinger SJ, Feldman EC, eds. Textbook of Veterinary Internal 506−508. Medicine. St. Louis, Missouri: Elsevier Saunders, 2005:1145−1165. 2. Laste NJ, Harpster NK. A retrospective study of 100 cats with 16. Drost WT, Bahr RJ, Henry GA, Campbell GA. Aortoiliac throm- feline distal aortic thromboembolism: 1977–1993. J Am Anim bus secondary to a mineralized arteriosclerotic lesion. Vet Radiol Hosp Assoc 1995;31:492−500. Ultrasound 1999;40:262−266. 3. Kittleson MD. Feline myocardial disease. In: Ettinger SJ, Feldman 17. Font A, Closa J. Ultrasonographic localization of a caudal EC, eds. Textbook of Veterinary Internal Medicine. St. Louis, vena cava thrombus in a dog with leishmaniasis. Vet Radiol Missouri: Elsevier Saunders, 2005:1082−1104. Ultrasound 1997;38:394−396. 4. Smith SA, Tobias AH, Jacob KA, Fine DM, Grumbles PL. Arterial 18. Lamb C. Doppler ultrasound examination in dogs and cats: thromboembolism in cats: Acute crisis in 127 cases (1992–2001) 2. Abdominal applications. In Practice 2005;27:238−247. and long-term management with low-dose aspirin in 24 cases. 19. Szatmàri V, Harkaanyi Z, Vöaröas K. A review of nonconven- J Vet Intern Med 2003;17:73−83. tional ultrasound techniques and contrast-enhanced ultrasonog- 5. Bekeredjian R, Grayburn PA, Shohet RV. Use of ultrasound raphy of noncardiac canine disorders. Vet Radiol Ultrasound contrast agents for gene or drug delivery in cardiovascular 2003;44:380−391. medicine. J Am Coll Cardiol 2005;45:329−335. 6. Li T, Tachibana K, Kuroki M, Kuroki M. Gene transfer with echo-enhanced contrast agents: Comparison between Albunex,

2000;64:0–00 The Canadian Journal of Veterinary Research 139 FOR PERSONAL USE ONLY Short Communication Communication brève

Seroprevalence of feline immunodeficiency virus (FIV) and feline leukemia virus (FeLV) in shelter cats on the island of Newfoundland, Canada Hannah J. Munro, Lesley Berghuis, Andrew S. Lang, Laura Rogers, Hugh Whitney

Abstract Feline immunodeficiency virus (FIV) and feline leukemia virus (FeLV) are retroviruses found within domestic and wild cat populations. These viruses cause severe illnesses that eventually lead to death. Housing cats communally for long periods of time makes shelters at high risk for virus transmission among cats. We tested 548 cats from 5 different sites across the island of Newfoundland for FIV and FeLV. The overall seroprevalence was 2.2% and 6.2% for FIV and FeLV, respectively. Two sites had significantly higher seroprevalence of FeLV infection than the other 3 sites. Analysis of sequences from the FeLV env gene (envelope gene) from 6 positive cats showed that 4 fell within the FeLV subtype-A, while 2 sequences were most closely related to FeLV subtype-B and endogenous feline leukemia virus (en FeLV). Varying seroprevalence and the variation in sequences at different sites demonstrate that some shelters are at greater risk of FeLV infections and recombination can occur at sites of high seroprevalence.

Résumé Le virus de l’immunodéficience féline (FIV) et le virus de la leucémie féline (FeLV) sont des rétrovirus retrouvés chez les populations de chats domestiques et sauvages. Ces virus causent des maladies sévères qui éventuellement mènent à la mort. L’hébergement de chats de façon communautaire pendant de longues périodes rend les refuges à risque élevé pour la transmission du virus parmi les chats. Nous avons testé 548 chats provenant de cinq sites différents à travers l’ile de Terre-Neuve pour FIV et FeLV. La séroprévalence globale était de 2,2 % et 6,2 % pour FIV et FeLV, respectivement. Deux sites avaient une séroprévalence significativement plus élevée d’infection par FeLV que les trois autres sites. L’analyse des séquences du gène env de FeLV (gène de l’enveloppe) provenant de six chats positifs a montré que quatre appartenaient au sous-type A de FeLV, alors que deux séquences étaient plus apparentées au sous-type B de FeLV et du virus endogène de la leucémie féline (en FeLV). Une séroprévalence variable et la variation dans les séquences à différents sites démontrent que certains refuges sont à risque plus élevé d’infections par FeLV et que de la recombinaison peut survenir aux sites avec une séroprévalence élevée. (Traduit par Docteur Serge Messier)

Feline immunodeficiency virus (FIV) and feline leukemia virus the mother to her developing offspring during pregnancy or to (FeLV) cause infectious diseases in both domestic cats (Felis catus) nursing young through infected milk (2). Transmission is most and wild cat species around the world. Both viruses can be transmit- successful when there is direct contact between cats because the ted horizontally, through saliva or other body fluids (1), and verti- virus is vulnerable to disinfectants, heat, and dry conditions (2). cal transmission probably also occurs (2,3). Both viruses are in the Infection with FeLV can result in impaired bone marrow func- family Retroviridae that is responsible for equine infectious anemia, tion and the development of certain forms of cancer, including caprine arthritis encephalitis, and many other diseases (4). These lymphomas (2). viruses have positive-sense, single-stranded ribonucleic acid (RNA) Feline immunodeficiency virus (FIV) is transferred from cat to cat genomes that integrate into the host genome after reverse transcrip- primarily by saliva through deep penetrating bites (3). The penetrat- tion to double-stranded deoxyribonucleic acid (DNA) upon infection. ing bites facilitate transmission by delivering the virus to the blood, They chronically infect immune system cells and their replication thereby bypassing the opportunity for the innate immune response can compromise the host’s immune system (2,3). to contain the virus at the site of injury. Acutely infected mothers As FeLV is mainly shed oronasally in saliva and tears as well as may transmit the virus to their developing offspring during preg- in the feces and urine (5), activities such as biting, mutual groom- nancy or through infected milk when nursing young (7). However, ing, and sharing food bowls and litter pans are modes of transmis- there is evidence of high levels of FIV RNA in the supernatant of sion (2,6). The virus, or antibodies, may also be transmitted from milk from an infected cat (3).

Department of Biology, Memorial University of Newfoundland, St. John’s, Newfoundland and Labrador A1B 3X9 (Munro, Berghuis, Lang); Animal Health Division, Department of Natural Resources, P.O. Box 7400, St. John’s, Newfoundland and Labrador A1E 3Y5 (Rogers, Whitney). Address all correspondence to Dr. Hugh Whitney; telephone: (709) 729-6879; fax: (709) 729-0055; e-mail: [email protected] Received April 9, 2013. Accepted June 13, 2013.

140 The Canadian Journal of Veterinary Research 2014;78:140–144 FOR PERSONAL USE ONLY

Figure 1. Phylogeny of FeLV env sequences. Branches labeled with filled circles represent sequences from the Newfoundland site with high seroprevalence of FeLV and the branch labeled with a filled square represents the sequence from the low seroprevalence site. Sub-types, as defined in Lutz et al (1), are indicated on the right, with the star indicating Feline Sarcoma Virus. The tree was constructed using the maximum likelihood method. Bootstrap support values based on 10 000 replicates are given at the nodes, with branches having , 50% support collapsed. Murine leukemia virus sequences (accession numbers RMU94692 and Z11128) were used as the outgroup.

The purposes of this study were: 1) to determine the sero­ tory, health, and breed were tested. This work was carried out under prevalence of FeLV and FIV infections in shelter cat populations in Protocol 09/10/HW from the Memorial University Institutional Newfoundland, and 2) to analyze the FeLV sequences found in cats Animal Care Committee to HW and Biosafety Permit # S-103-1-08 from Newfoundland. This is the first survey to focus on FIV and to ASL. FeLV in animal shelters on the island of Newfoundland. A veterinary professional obtained blood samples from each cat A total of 548 cats was tested for FIV and FeLV infection. Of these, either through intravenous or intracardiac extraction. Blood taken 366 samples were obtained from the St. John’s Humane Services, 112 through intracardiac extraction was from heavily sedated cats that from the St. John’s Society for the Prevention of Cruelty to Animals were euthanized immediately after blood collection. The blood (SPCA), 19 from the Clarenville SPCA, and 51 from 2 shelters in was put into tubes either with EDTA anticoagulant or without western Newfoundland. Cats of various ages, gender, ownership his- anticoagulant from which the serum was removed and frozen. Each

2000;64:0–00 The Canadian Journal of Veterinary Research 141 FOR PERSONAL USE ONLY

Table I. Occurrence of FIV- and FeLV-seropositive cats in 5 shelters in Newfoundland. P-values from general linear model (GLM) analysis comparing infection status (FIV or FeLV) and shelter, age, and sex are shown Number of cats FIV FeLV tested 1 (%) P-value 1 (%) P-value Shelter 0.46 . 0.01 Site 1 366 10 (2.7) 8 (2.2) Site 2 112 2 (1.8) 3 (2.7) Site 3 19 0 0 Site 4 35 0 14 (40) Site 5 16 0 9 (56.3) Age 0.02 0.01 Young kitten 135 1 (0.7) 0 Old kitten 39 0 0 Young adult 39 0 1 (2.6) Adult 263 9 (3.4) 22 (8.4) Old adult 14 1 (7.1) 1 (7.1) Geriatric 26 1 (3.9) 1 (3.8) Unknown 32 0 9 (28.1) Gender . 0.01 0.23 Females 258 1 (0.4) 10 (3.9) Males 261 11 (4.2) 15 (5.8) Unknown 29 0 9 (31.0) Total 548 2.2 6.2

sample was tested for FIV and FeLV using the IDEXX SNAP FIV/ (5 to 15 ng/mL), 0.125 μL of 0.1 U/μL Taq DNA Polymerase (NEB, FeLV Combo Test Diagnostic Kit (IDEXX Laboratories, Markham, Oakville, Ontario), 2.5 μL of 103 ThermoPol reaction buffer, 0.5 μL of Ontario). This kit uses enzyme-linked immunosorbent assay (ELISA) each 10-μM primer stock, and 0.5 μL of 10-mM dNTPs. Polymerase technology to detect the presence of the p27 antigen from FeLV and chain reaction (PCR) thermal cycling conditions were an initial antibodies to the p24 antigen of FIV (8). denaturing step of 95°C for 5 min, followed by 35 cycles of 95°C Statistical analysis was done in R (9). A generalized linear model denaturing for 30 s, 57°C annealing for 30 s, and 72°C extension for was fitted with a binomial distribution and logit-link function. 2 min, and a final extension at 72°C for 5 min. Products of PCR were Feline leukemia virus (FeLV) and FIV were analyzed separately with visualized by gel electrophoresis on a 1% agarose gel and samples infection status as the dependent variable and shelter, gender, and with a single 2.1-kbp amplicon were purified with the QIAquick PCR age as the independent variables. Cats were organized into 6 age Purification Kit (Qiagen) before being sequenced with the env primer groups: young kitten — birth to 11 wk; older kitten — 3 to 5 mo; (10). For 3 of the samples due to poor sequencing results, the PCR young adult — 6 to 11 mo; adult — 1 to 6 y; older adult — 7 to 10 y; product was cleaned and subsequently cloned using the pGEM-T and geriatric — 11 y and older. Easy Vector (Promega, Madison, Wisconsin, USA) in Escherichia coli Five FeLV p27 antigen-positive samples from 1 shelter location in NEB 5a competent cells (NEB) and sequenced from the resulting western Newfoundland and 1 from St. John’s Humane Services were plasmids. DNA sequencing was carried out at the Centre for Applied used for virus sequence analysis. Samples from the shelter in western Genomics (Toronto, Ontario). Raw sequence data were trimmed, Newfoundland were considered to come from a high seroprevalence with 726 base pair sequences from the env gene used for subsequent FeLV site, whereas the sample from the St. John’s Humane Services analyses. The sequences have been deposited in the GenBank data- was considered to come from a lower seroprevalence site. DNA was base of the National Center for Biotechnology Information (NCBI) extracted from blood samples with anticoagulant using the Qiagen under accession numbers KC540945-KC540950. Generation Capture Column Kit (Qiagen, Mississauga, Ontario) as The Newfoundland cat virus sequences were compared to known per the manufacturer’s recommendations. The DNA was then used domestic cat FeLV sequences from a range of strains and subtypes for polymerase chain reaction (PCR) with primers that amplify from (11), as indicated in Figure 1. Alignments and phylogenetic analyses within the env (envelope) gene to the end of the LTR14 region (10). were carried out with MEGA5.05 (12). The nucleotide sequences We modified the PCR conditions from those previously published were aligned using multiple sequence alignment (MUSCLE) (13) and (10) as follows: the 25-mL reactions contained 1 mL template DNA the evolutionary history was inferred using the maximum likelihood

142 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

method based on the Hasegawa-Kishino-Yano model (14). There disease transmission through bites and fighting. This highlights the were a total of 726 nucleotide positions in the final dataset and the importance of regularly testing cats in shelters with high densities of phylogeny was bootstrapped with 10 000 replicates. cats, maintaining proper sanitation, and following guidelines such In total, 2.2% of the cats tested seropositive for FIV and 6.2% as those set out by the American Association of Feline Practitioners tested positive for FeLV. The seroprevalence of FIV, but not of FeLV, (8,19). Two of the sites in this study have a high turnover rate, i.e., varied significantly between genders (Table I). Infection with FeLV, cats usually do not stay at either shelter for more than a few months. but not with FIV, varied among shelters sampled, with 2 sites having This limited amount of time of exposure to any asymptomatic cats significantly higher levels of FeLV (Table I). Older cats were more may result in the lower prevalence. likely to be positive for either FIV or FeLV. In conclusion, male cats seem to be more at risk for both viruses, A portion of the FeLV genome (part of the env gene) was ampli- especially FIV. Also, those males that tested positive were mainly fied by PCR and sequenced from 6 of the positive samples. This intact. There is a link between shelter management styles charac- included samples from 2 different locations. The resulting sequences terized by maintaining a low density cat population, isolating sick were compared to previously reported sequences in the GenBank animals, paying stringent attention to proper hygiene, and having database by using the basic local alignment search tool (BLAST). a higher turnover rate and a relatively lower prevalence of FIV Three of the 5 samples from the high seroprevalence site were very and FeLV in the shelter animals. Also, for those shelters that are similar (. 99% identity) to each other. These 3 sequences were interested in reducing the prevalence of FIV and FeLV in the cat closely related (. 99% identity) to 945 env, a well characterized and population and adopting out healthy cats to prospective owners, highly virulent strain (15) within the FeLV subtype-A (FeLV-A). a second FIV/FeLV test would be beneficial when cats are held for The sequence from the low seroprevalence site was also related long periods of time or have higher risk of contact with the disease, to strains within FeLV-A, as well as being closely related to feline given the timeline of progression of these viral diseases. sarcoma virus, which groups within the clade containing FeLV-A. Feline sarcoma virus is a product of recombination of a FeLV-A Acknowledgments virus with cellular oncogenes (16). The other 2 sequences from the high seroprevalence site are most closely related to endogenous The authors thank Cathy Keane, Robin Janes, Heather Quilty, proviruses and FeLV subtype-B (FeLV-B). Feline leukemia virus Donna Scott, Dr. Beverly Dawe, Dr. Vicki O’Leary, Debbie Powers, subtype B (FeLV-B) originated from recombination of FeLV-A with Susan Deir, Cindy McGrath, and Candace King for help with the endogenous FeLV (enFeLV) (2). A phylogenetic analysis supported collection of samples and this study. We thank IDEXX Laboratories these BLAST results and showed that the Newfoundland sequences for providing the kits for sample screening and acknowledge were not monophyletic (Figure 1). the support of Jonas Goring and Paul Day (Pro-Medix). Hannah The seroprevalence of FIV and FeLV varies among shelter popu- Munro was supported in part by a fellowship from the School of lations. Past studies have found averages as low as 1.7% (17) and Graduate Studies at Memorial University. Lesley Berghuis’ research as high as 6.4% (18) for FIV in shelters throughout Canada. Within was supported in part by funding from the Memorial University the Province of Newfoundland and Labrador in both shelter and Department of Biology Honours program. Funding was provided by privately owned cats, the seroprevalence of FeLV and FIV was the Newfoundland and Labrador Department of Natural Resources previously found to be 4.33% and 5.00%, respectively (18). Of the and a Discovery Grant to Andrew Lang from the Natural Sciences 5 shelters we sampled, only 2 had cats that tested positive for FIV. and Engineering Research Council (NSERC) of Canada funded the One shelter had a seroprevalence of 1.8% and the other 2.7%, which virus sequencing. is well within previous findings in Canada. Similar to prevalence of FIV, FeLV has ranged from 1.5% (8) to 2.7%, depending on the References study (18). Four of the 5 sites that we sampled had 1 or more FeLV- positive cats and the seroprevalence ranged from 2.2% to 56.3%. The 1. Lutz H, Addie D, Belak S, et al. Feline leukaemia. ABCD remarkably high percentage of positives was found at a site with guidelines on prevention and management. J Feline Med Surg a high-density cat population and poor sanitation, which further 2009;11:565–574. supports the need for proper hygienic practices to limit the spread 2. Hartmann K. Feline leukemia virus infection. In: Greene CE, of FeLV within shelters through testing, prevention, and responding ed. Infectious Diseases of the Dog and Cat. 3rd ed. St. Louis, appropriately to positive cats. Missouri: Saunders, 2006:105–131. Preventative methods are usually high in most local shelters 3. Sellon R, Hartmann K. Feline immunodeficiency virus infection. due to the routine testing of cats and strict protocols for cleaning In: Greene CE, ed. Infectious Diseases of the Dog and Cat. 3rd ed. potentially contaminated surfaces and equipment (8). However, St. Louis, Missouri: Saunders, 2006:131–142. FeLV can easily spread through an enclosed population because it 4. Stoye JP, Blomberg J, Coffin JM, et al. Retroviridae. In: King AMQ, is readily transmitted among cats and it takes up to 30 d for ELISA- Adams MJ, Carstens EB, Lefkowitz EJ, eds. Virus Taxonomy: based tests to accurately diagnose infection (8), which delays the Classification and Nomenclature of Viruses. St. Louis, Missouri: ability to detect and react to infection. Another factor that contributes Elsevier, 2011:477–495. to spread of the virus is the level of isolation of the cats from one 5. Jarrett W, Jarrett O, Mackey L, Laird H, Hardy W, Essex M. another. Distressed, aggressive, and sick cats are usually kept in Horizontal transmission of leukemia virus and leukemia in the cages that separate them from the others, reducing the chances of cat. J Natl Cancer Inst 1973;51:833–841.

2000;64:0–00 The Canadian Journal of Veterinary Research 143 FOR PERSONAL USE ONLY

6. Gomes-Keller MA, Gönczi E, Tandon R, et al. Detection of feline 14. Hasegawa M, Kishino H, Yano T. Dating of the human-ape leukemia virus RNA in saliva from naturally infected cats and ­splitting by a molecular clock of mitochondrial DNA. J Mol Evol correlation of PCR results with those of current diagnostic meth- 1985;22:160–174. ods. J Clin Microbiol 2006;44:916–922. 15. Levesque KS, Bonham L, Levy LS. flvi-1, a common integration 7. O’Neil LL, Burkhard MJ, Hoover EA. Frequent perinatal trans- domain of feline leukemia virus in naturally occurring lympho- mission of feline immunodeficiency virus by chronically infected mas of a particular type. J Virol 1990;64:3455–3462. cats. J Virol 1996;70:2894–2901. 16. Donner L, Fedele LA, Garon CF, Anderson SJ, Sherr CJ. 8. Levy J, Crawford C, Hartmann K, et al. 2008 American McDonough feline sarcoma virus: Characterization of the molec- Association of Feline Practitioners’ feline retrovirus manage- ularly cloned provirus and its feline oncogene (v-fms). J Virol ment guidelines. J Feline Med Surg 2008;10:300–316. 1982;41:489–500. 9. R Core Team. R: A Language and Environment for Statistical 17. Levy JK, Scott HM, Lachtara JL, Crawford PC. Seroprevalence Computing. Vienna, Austria: R Foundation for Statistical of feline leukemia virus and feline immunodeficiency virus Computing; 2012. infection among cats in North America and risk factors for 10. Athas GB, Choi B, Prabhu S, Lobelle-Rich PA, Levy LS. Genetic seropositivity. J Am Vet Med Assoc 2006;228:371–376. determinants of feline leukemia virus-induced multicentric 18. Little S, Sears W, Lachtara J, Bienzle D. Seroprevalence of feline lymphomas. Virology 1995;214:431–438. leukemia virus and feline immunodeficiency virus infection 11. Coelho FM, Bomfim MR, de Andrade Caxito F, et al. Naturally among cats in Canada. Can Vet J 2009;50:644–648. occurring feline leukemia virus subgroup A and B infections in 19. Little S, Bienzle D, Carioto L, Chisholm H, O’Brien E, Scherk M. urban domestic cats. J Gen Virol 2008;89:2799–2805. Feline leukemia virus and feline immunodeficiency virus in 12. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S. Canada: Recommendations for testing and management. Can MEGA5: Molecular evolutionary genetics analysis using maxi- Vet J 2011;52:849–855. mum likelihood, evolutionary distance, and maximum parsi- mony methods. Mol Biol Evol 2011;28:2731–2739. 13. Edgar RC. MUSCLE: Multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res 2004;32: 1792–1797.

144 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY Short Communication Communication brève

Antimicrobial susceptibility of Streptococcus suis isolated from clinically healthy swine in Brazil Taíssa Cook Siqueira Soares, Antonio Carlos Paes, Jane Megid, Paulo Eduardo Martins Ribolla, Karina dos Santos Paduan, Marcelo Gottschalk

Abstract Streptococcus suis is an important pathogen in the swine industry. This study is the first to report on the antimicrobial susceptibility of S. suis isolated from clinically healthy pigs in Brazil; the fourth major pork producer in the world. The antimicrobial susceptibility of 260 strains was determined by disc diffusion method. Strains were commonly susceptible to ceftiofur, cephalexin, chloramphenicol, and florfenicol, with more than 80% of the strains being susceptible to these antimicrobials. A high frequency of resistance to some of the antimicrobial agents was demonstrated, with resistance being most common to sulfa-trimethoprim (100%), tetracycline (97.69%), clindamycin (84.61%), norfloxacin (76.92%), and ciprofloxacin (61.15%). A high percentage of multidrug resistant strains (99.61%) were also found. The results of this study indicate that ceftiofur, cephalexin, and florfenicol are the antimicrobials of choice for empirical control of the infections caused by S. suis.

Résumé Streptococcus suis est un pathogène important de l’industrie porcine. Cette étude est la première à rapporter la susceptibilité antimicrobienne de souches S. suis isolées de porcs cliniquement sains provenant du Brésil, le quatrième producteur de porc à l’échelle de la planète. La susceptibilité antimicrobienne de 260 souches fut examinée par la méthode de diffusion des disques imprégnés d’antibiotiques. Plusieurs souches étaient susceptibles au ceftiofur, céphalexine, chlorphénicol, et florfénicol, avec plus de 80 % des souches qui étaient susceptibles à ces antimicrobiens. Une haute fréquence de résistance à certains des antimicrobiens futs démontrée, principalement au sulfa-triméthoprime (100 %), tetracycline (97,69 %), clindamycine (84,61 %), norfloxacine (76,92 %), et ciprofloxacine (61,15 %). Un pourcentage élevé de souches multirésistantes (99,61 %) a également été observé. Les résultats de cette étude indiquent que le ceftiofur, la céphalexine et le florfénicol sont les antimicrobiens de choix pour le contrôle empirique des infections causées par S. suis. (Traduit par les auteurs)

Streptococcus suis is recognized worldwide as an important patho- countries (8–12). Differences in the level of resistance have been gen in intensive swine production and has been associated with observed between different countries, serotypes, and over time meningitis, septicemia, endocarditis, arthritis, pneumonia, and, occa- (9,13,14). In Brazil, there is only one report showing the susceptibility sionally, endometritis, abortion, rhinitis, and vaginitis (1). Moreover, profiles of strains of S. suis isolated from diseased pigs (15). Brazil is S. suis is a zoonotic agent and may cause a variety of infections in the fourth major pork producer in the world, producing, consuming, individuals working in close contact with swine or pork products and exporting around 3300, 2700, and 600 million tonnes, respec- (2). Strains of S. suis are divided into 35 serotypes, according to its tively. The objective of this study was to determine the antimicrobial polysaccharide capsular antigens (3–6). The natural habitat of S. suis susceptibility, as well as the resistotypes, of Brazilian S. suis strains is the upper respiratory tract, particularly the tonsils and nasal cavi- recovered from clinically healthy pigs. ties, and the genital and alimentary tracts of pigs. The movement of A total of 260 S. suis field strains were included in this study. healthy carrier pigs harboring the microorganism is the main route During a 4-month period from November 2009 to February 2010, of transmission between herds (7). palatine tonsils, nasal cavities, and vaginal samples were collected Since current vaccines are bacterins and provide only partial using cotton swabs from 240 clinically healthy pigs, including serotype-specific protection, antimicrobial agents have become 24 sows, 96 suckling pigs, 60 nursery pigs, and 60 fattening pigs, increasingly important in treating and controlling infection caused from 3 different farms located in 3 different cities within the state by S. suis. Previous studies have investigated the antimicrobial of São Paulo, Brazil; one of the most important swine producing susceptibility of strains from diseased and healthy pigs in different states in Brazil.

Departamento de Higiene Veterinária e Saúde Pública. Universidade Estadual Paulista Júlio de Mesquita Filho, São Paulo, Brazil (Cook Siqueira Soares, Paes, Megid); Faculty of Veterinary Medicine, Université de Montréal, Montreal, Quebec (Gottschalk); and Departamento de Parasitologia. Universidade Estadual Paulista Júlio de Mesquita Filho, São Paulo, Brazil (Martins Ribolla, dos Santos Paduan). Address all correspondence to Dr. Marcelo Gottschalk; telephone: 450-773-8521, ext. 8374; fax: 450-778-8108; e-mail: [email protected] Dr. Gottschalk’s current address is Université de Montréal, Faculté de médecine vétérinaire/College of Veterinary Medicine, 3200 Sicotte, ­Saint-Hyacinthe, Quebec J2S 2M2. Received March 22, 2013. Accepted July 4, 2013.

2014;78:145–149 The Canadian Journal of Veterinary Research 145 FOR PERSONAL USE ONLY

Table I. Specific breakpoints for each of the antibiotics used Table II. Antimicrobial susceptibility of 260 Streptococcus suis strains isolated from clinically normal pigs in Brazil Antimicrobials Susceptible Intermediate Resistant Ampicillin ≥ 26 19–25 # 18 Susceptible Intermediate Resistant Azithromycin ≥ 18 14–17 # 13 Antimicrobials strains (%) strains (%) strains (%) Cephalexin ≥ 18 15–17 # 14 Ampicillin 174 (66.92) 69 (26.54) 17 (6.54) Ceftiofur ≥ 21 18–20 # 17 Azithromycin 189 (72.69) 22 (8.46) 49 (18.85) Ciprofloxacin ≥ 21 16–20 # 15 Cephalexin 219 (84.23) 4 (1.54) 37 (14.23) Clindamycin ≥ 19 16–18 # 15 Ceftiofur 248 (95.39) 9 (3.46) 3 (1.15) Chloramphenicol ≥ 21 18–20 # 17 Ciprofloxacin 64 (24.62) 37 (14.23) 159 (61.15) Doxycycline ≥ 16 13–15 # 12 Clindamycin 6 (2.31) 34 (13.08) 220 (84.61) Erythromycin ≥ 21 16–20 # 15 Chloramphenicol 213 (81.92) 25 (9.62) 22 (8.46) Enrofloxacin ≥ 23 17–22 # 16 Doxycycline 162 (62.31) 70 (26.92) 28 (10.77) Florfenicol ≥ 22 19–21 # 18 Erythromycin 124 (47.69) 15 (5.77) 121 (46.54) Levofloxacin ≥ 17 14–16 # 13 Enrofloxacin 115 (44.23) 31 (11.92) 114 (43.85) Norfloxacin ≥ 17 13–16 # 12 Florfenicol 212 (81.54) 10 (3.85) 38 (14.61) Penicillin ≥ 28 22–27 # 21 Levofloxacin 162 (62.31) 16 (6.15) 82 (31.54) Sulfa 1 Trimethoprim ≥ 16 11–15 # 10 Norfloxacin 46 (17.69) 14 (5.39) 200 (76.92) Tetracycline ≥ 23 19–22 # 18 Penicillin 127 (48.85) 86 (33.08) 47 (18.07) Sulfa 1 Trimethoprim 0 (0.00) 0 (0.00) 260 (100.00) Tetracycline 6 (2.31) 0 (0.00) 254 (97.69)

Swabs were cultured on a 5% sheep blood agar plate. The plates were incubated aerobically at 37°C and inspected for growth after 24 and 48 h. Three colonies 1 to 2 mm in diameter, showing alpha- more serotypes, which is highly common within isolates from clini- hemolysis, were suspected as potential S. suis (16). The S. suis-like cally healthy animals (unpublished observations). strains were identified using polymerase chain reaction (PCR). The The results of the susceptibility testing of the S. suis strains are DNA extraction was done using the Chelating Ion Exchange Resin shown in Table II. No significant differences were observed when (Chelex-100 resin; Bio-Rad Laboratories, California, USA) and the comparing isolates recovered from different animal categories or amplification was done as described by Okwumabua et al (17). The regions of the country (result not shown). More than 80% of the PCR positive strains were serotyped by coagglutination technique strains were susceptible to ceftiofur (95.39%), cephalexin (84.23%), with sera against the 34 described serotypes (18). chloramphenicol (81.92%), and florfenicol (81.54%). A high frequency The antimicrobial susceptibility of S. suis strains was determined of resistance to some of the antimicrobial agents was demonstrated, by using the disc diffusion method in Mueller-Hinton agar supple- with resistance being most common to sulfa-trimethoprim (100%), mented with 5% defibrinated sheep blood, according to the recom- tetracycline (97.69%), clindamycin (84.61%), norfloxacin (76.92%), mendation by CLSI (19), following specific breakpoints indicated and ciprofloxacin (61.15%). by the manufacturer (Cefar Diagnóstica Ltda, São Paulo, Brazil) as The antimicrobial susceptibility profiles (resistotypes) of the S. suis shown in Table I. The antimicrobial agents used in this study were: strains studied were constructed using all 16 antimicrobials agents. ampicillin (10 μg); ceftiofur (30 μg); ciprofloxacin (5 μg); florfenicol The overall distribution of resistotypes is shown in Table III. None (30 μg); chloramphenicol (30 μg); norfloxacin (10 μg); enrofloxacin of the 260 S. suis strains was susceptible to the 16 antimicrobials. (5 μg); sulfa-trimethoprim (25 μg); tetracycline (30 μg); levofloxacin Multidrug resistant strains (≥ 3 antimicrobial agents) were observed (5 μg); doxycycline (30 μg); and penicillin (10 UI), erythromycin in 99.61% (259/260) of all isolates. Two hundred and fifty-three (15 μg), cephalexin (30 μg), azithromycin (15 μg), and clindamycin strains (97.30%) were resistant to at least 4 antimicrobials, 221 strains (2 μg). The chi-squared test was used to estimate the relationship (85.00%) were resistant to at least 6 antimicrobials, and 9 strains were among herd, animal category, and antimicrobial susceptibility. resistant to all 16 drugs. Of the 59 resistotypes, resistotype 19 (ampi- Differences were considered significant when P , 0.05. cillin, ciprofloxacin, clindamycin, doxycycline, norfloxacin, penicil- Results from serotyping showed that most isolates (67%) were lin, sulfa-trimethoprim, and tetracycline), with 28 strains (10.76%), untypable with only 86 isolates being serotyped. The fact that most was the most frequently isolated. The strains belonging to serotype 5 strains isolated from clinically healthy animals are untypable is a were resistant to 11 different antimicrobials; those of serotype 7 were common finding (M. Gottschalk, reference laboratory for S. suis sero- resistant to 9 or 10 antimicrobials; those of serotype 19 were resistant typing, unpublished data). Among the typable isolates, serotypes 22 to 9 different antimicrobials; those of serotype 21 were resistant to 7, (16 isolates), 30 (13 isolates), 21 (10 isolates), and 27 (9 isolates) were 8, 9, or 10 (7 to 10) different antimicrobials; and those of serotype 11 the most commonly identified. Serotypes 5, 7, 11, 14, 15, 16, 18, 19, were resistant to 4, 5, 6, or 7 (4 to 7) different antimicrobials. The 20, 24, 28, 29, and 34 were also identified with less than 3 isolates/ strains that were important to public health were resistant to 7 (sero- serotypes. All of these serotypes have already been described as type 14), 10 (serotype 16), and 13 to 14 (serotype 24) antimicrobials. causes of disease in pigs (20) and, some of them (serotypes 5, 14, A high frequency of resistance to tetracyclines (tetracycline/ 16, and 24), in humans (2,21). A total of 16 isolates reacted with 2 or 97.69%), sulphonamides (sulfa-trimethoprim/100%), macrolides

146 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

Table III. Resistotypes of 260 Streptococcus suis strains isolated from clinically normal pigs in Brazil Number of Number of antimicrobials strains Resistotype 2 1 CLI-SULT 3 3 CLI-SULT-TET 3 3 ENR-SULT-TET 4 12 CLI-ERI-SULT-TET 5 1 AZI-CIP-ERI-ENR-SULT 5 1 AZI-CLI-ERI-SULT-TET 5 2 AZI-CLI-DOX-SULT-TET 5 14 CLI-ERI-NOR-SULT-TET 5 2 ENR-LEV-NOR-SULT-TET 6 6 AZI-CLI-ERI-NOR-SULT-TET 6 4 AZI-CLI-ERI-PEN-SULT-TET 7 4 AMP-CIP-CLI-NOR-PEN-SULT-TET 7 5 AZI-CIP-CLI-ERI-ENR-SULT-TET 7 3 AZI-CIP-CLI-ERI-NOR-SULT-TET 7 2 AZI-CIP-CLI-ERI-PEN-SULT-TET 7 1 AZI-CLI-ERI-NOR-PEN-SULT-TET 7 1 AZI-CLI-ERI-ENR-NOR-SULT-TET 7 7 CIP-CLI-ERI-ENR-NOR-SULT-TET 8 28 AMP-CIP-CLI-DOX-NOR-PEN-SULT-TET 8 11 AZI-CIP-CLI-CLO-ERI-NOR-SULT-TET 8 5 AZI-CIP-CLI-ERI-ENR-NOR-SULT-TET 8 2 AZI-CIP-CLI-ERI-NOR-PEN-SULT-TET 8 3 AZI-CIP-CLI-DOX-ERI-NOR-SULT-TET 8 5 CIP-CLI-ERI-ENR-LEV-NOR-SULT-TET 9 3 AMP-AZI-CIP-CLI-ERI-ENR-NOR-PEN-SULT 9 6 AMP-CIP-CLI-DOX-ENR-NOR-PEN-SULT-TET 9 2 AZI-CEFA-CIP-CLI-ERI-ENR-NOR-SULT-TET 9 13 AZI-CIP-CLI-ERI-ENR-LEV-NOR-SULT-TET 9 2 AZI-CIP-CLI-ERI-ENR-NOR-PEN-SULT-TET 9 4 AZI-CIP-CLI-DOX-ERI-ENR-NOR-SULT-TET 10 12 AMP-AZI-CEFA-CLI-CLO-ERI-FLO-PEN-SULT-TET 10 2 AMP-AZI-CIP-CLI-ERI-ENR-NOR-PEN-SULT-TET 10 3 AZI-CEFA-CIP-CLI-DOX-ERI-ENR-NOR-SULT-TET 10 3 AZI-CIP-CLI-CLO-DOX-ERI-FLO-NOR-SULT-TET 10 11 AZI-CIP-CLI-ERI-ENR-LEV-NOR-PEN-SULT-TET 10 11 AZI-CIP-CLI-DOX-ERI-ENR-LEV-NOR-SULT-TET 11 1 AMP-AZI-CIP-CLI-DOX-ERI-ENR-LEV-NOR-SULT-TET 11 4 AMP-AZI-CIP-CLI-ERI-ENR-LEV-NOR-PEN-SULT-TET 11 2 AZI-CEFA-CIP-CLI-ERI-ENR-LEV-NOR-PEN-SULT-TET 11 5 AZI-CIP-CLI-ERI-ENR-FLO-LEV-NOR-PEN-SULT-TET 11 4 AZI-CIP-CLI-DOX-ERI-ENR-FLO-LEV-NOR-SULT-TET 11 5 AZI-CIP-CLI-DOX-ERI-ENR-LEV-NOR-PEN-SULT-TET 12 1 AMP-AZI-CEFA-CIP-CLI-DOX-ERI-ENR-NOR-PEN-SULT-TET 12 1 AMP-AZI-CEFA-CIP-CLI-ERI-ENR-LEV-NOR-PEN-SULT-TET 12 3 AMP-AZI-CIP-CLI-DOX-ERI-ENR-FLO-NOR-PEN-SULT-TET 12 2 AMP-AZI-CIP-CLI-DOX-ERI-ENR-LEV-NOR-PEN-SULT-TET 12 2 AMP-AZI-CIP-CLI-ERI-ENR-FLO-LEV-NOR-PEN-SULT-TET 12 4 AZI-CIP-CLI-CLO-DOX-ERI-ENR-LEV-NOR-PEN-SULT-TET 12 2 AZI-CIP-CLI-CLO-DOX-ERI-ENR-FLO-LEV-NOR-SULT-TET 12 1 AZI-CIP-CLI-DOX-ERI-ENR-FLO-LEV-NOR-PEN-SULT-TET 13 2 AMP-AZI-CEFA-CEFT-CIP-CLI-ERI-ENR-LEV-NOR-PEN-SULT-TET 13 2 AMP-AZI-CEFA-CIP-CLI-DOX-ERI-ENR-LEV-NOR-PEN-SULT-TET

2000;64:0–00 The Canadian Journal of Veterinary Research 147 FOR PERSONAL USE ONLY

Table III. (continued) 13 3 AMP-AZI-CIP-CLI-DOX-ERI-ENR-FLO-LEV-NOR-PEN-SULT-TET 13 3 AZI-CEFA-CIP-CLI-DOX-ERI-ENR-FLO-LEV-NOR-PEN-SULT-TET 13 1 AZI-CIP-CLI-CLO-DOX-ERI-ENR-FLO-LEV-NOR-PEN-SULT-TET 14 1 AMP-AZI-CEFT-CIP-CLI-CLO-DOX-ERI-ENR-FLO-NOR-PEN-SULT-TET 14 4 AZI-CEFA-CIP-CLI-CLO-DOX-ERI-ENR-FLO-LEV-NOR-PEN-SULT-TET 16 6 AMP-AZI-CEFA-CEFT-CIP-CLI-CLO-DOX-ERI-ENR-FLO-LEV-NOR-PEN-SULT-TET 16 3 AMP-AZI-CEFA-CEFT-CIP-CLI-CLO-DOX-ERI-ENR-FLO-LEV-NOR-PEN-SULT-TET AMP — Ampicillin; AZI — Azithromycin; CEFA — Cephalexin; CEFT — Ceftiofur; CIP — Ciprofloxacin; CLI — Clindamycin; CLO — Chloramphenicol; DOX — Doxycycline; ERI — Erythromycin; ENR — Enrofloxacin; FLO — Florfenicol; LEV — Levofloxacin; NOR — Norfloxacin; PEN — Penicillin; SULT — Sulfa 1 Trimethoprim; TET — Tetracycline.

(erythromycin/46.54%), and lincosamides (clindamycin/84.61%) was studies with isolates recovered over several years are necessary to found within the Brazilian strains. Streptococcus suis strains isolated evaluate the progression of resistant patterns within Brazilian strains. from diseased or clinically healthy pigs with high levels of resistance Beta-lactams are still the most active antimicrobials against S. suis to these classes of antimicrobials have already been described in the strains. The results of this study show that ceftiofur, cephalexin, and literature for other countries (9–11,22) and could be related with the florfenicol are the antimicrobials of choice for empirical control of wide use of these antimicrobials in veterinary medicine. the infections caused by S. suis in Brazil. The presence of resistant The antimicrobial agents representing the class of quinolones that and intermediate strains to ampicillin and penicillin also suggests showed high levels of resistance: norfloxacin (76.92%), ciprofloxa- the need for a continuous surveillance of the susceptibility pattern cin (61.15%), enrofloxacin (43.85%), and levofloxacin (31.54%). The of this pathogen. The high percentage of multidrug resistant strains resistance to enrofloxacin (43.85%) demonstrated in this study was belonging to serotypes previously described as causing clinical dis- higher than the results of strains from diseased pigs in Spain (2%) ease in pigs and humans, may lead to complications in the control (10) and other European countries (0%) (11). and treatment of the infections by S. suis. The result of this study showed a higher rate of resistance to flor- fenicol (14.61%) than to chloramphenicol (8.46%). Similarly, a higher Acknowledgments rate of resistance to florfenicol (21.1%) was demonstrated by Zhang et al (12). Since chloramphenicol has been banned in food animals This work was partially supported by the Fundação de Amparo à and replaced by florfenicol, antimicrobial susceptibility studies have Pesquisa do Estado de São Paulo (FAPESP) and the Canadian Bureau been showing an increase in the rate of resistance to florfenicol. for International Education. Among beta-lactams, the most effective antimicrobial was ceftio- fur (95.39%), followed by cephalexin (84.23%), ampicillin (66.92%), References and penicillin (48.85%). These results are in accordance with other studies that have described S. suis strains as mainly susceptible to 1. Staats JJ, Feder I, Okwumabua O, Chengappa MM. Streptococcus these antimicrobial agents (9,10,13,14). However, a large number of suis: Past and present. Vet Res Commun 1997;21:381–401. strains were classified as intermediate to ampicillin (26.54%) and to 2. Gottschalk M, Xu J, Calzas C, Segura M. Streptococcus suis: A new penicillin (33.08%). This indicates that there is increased antimicro- emerging or an old neglected zoonotic pathogen? Fut Microbiol bial resistance to this first line class of antimicrobial agents. 2010;5:371–391. The most prevalent resistotypes demonstrated by Vela et al (10) 3. Perch B, Pedersen KB, Henrichsen J. Serology of capsu- and Zhang et al (12) were formed by 4 and 5 antimicrobials, respec- lated streptococci pathogenic for pigs: Six new serotypes of tively, whereas the most prevalent resistotype found in this study Streptococcus suis. J Clin Microbiol 1983;17:993–996. consisted of 8 antimicrobials. The percentage of multidrug resistant 4. Gottschalk M, Higgins R, Jacques M, Mittal KR, Henrichsen J. strains showed in this study (99.61%) was higher than that demon- Description of 14 new capsular types of Streptococcus suis. J Clin strated by Zhang, et al (12), far beyond those described in studies Microbiol 1989;27:2633–2636. involving clinical strains (10,23), the highest already reported for 5. Gottschalk M, Higgins R, Jacques M, Beaudoin M, Henrichsen S. suis strains. The differences observed between countries might J. Characterization of six new capsular types (23 through 28) of be explained by different usage of antimicrobial agents or due to Streptococcus suis. J Clin Microbiol 1991;29:2590–2594. methodological differences between studies. Standardization of 6. Higgins R, Gottschalk M, Boudreau M, Lebrun A, Henrichsen J. susceptibility testing methods at an international level is necessary Description of six new capsular types (29–34) of Streptococcus to facilitate comparisons. suis. J Vet Diagn Invest 1995;7:405–406. In conclusion, this is the first study not only in Brazil, but also in 7. Clifton-Hadley FA, Alexander TLJ. The carrier rate and carrier the American continent, of antibiotic resistance from isolates recov- site of Streptococcus suis type II in pigs. Vet Rec 1980;107:40–41. ered from clinically healthy pigs. It is important to mention that 8. Han DU, Choi C, Ham HJ, et al. Prevalence, capsular type and results presented in this study represent isolates recovered in recent antimicrobial susceptibility of Streptococcus suis isolated from years in one important swine producing state in Brazil. Further slaughter pigs in Korea. Can J Vet Res 2001;65:151–155.

148 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

9. Marie J, Morvan H, Berthelot-Hérault F, et al. Antimicrobial 17. Okwumabua O, O’connor M, Shull E. A polymerase chain susceptibility of Streptococcus suis isolated from swine in France reaction (PCR) assay for Streptococcus suis based on the gene and from humans in different countries between 1996 and 2000. encoding the glutamate dehydrogenase. FEMS Microbiol Lett J Antimicrob Chemot 2002;50:201–209. 2003;218:79–84. 10. Vela AI, Moreno MA, Cebolla JA, et al. Antimicrobial susceptibil- 18. Gottschalk M, Higgins R, Bourdreau M. Use of polyvalent coag- ity of clinical strains of Streptococcus suis isolated from pigs in glutination reagents for serotyping of Streptococcus suis. J Clin Spain. Vet Microbiol 2005;105:143–147. Microbiol 1993;31:2192–2194. 11. Wisselink HJ, Veldman KT, Den Eede CV, Salmon SA, Mevius DJ. 19. Clinical and Laboratory Standards Institute. 2006. Performance Quantitative susceptibility of Streptococcus suis strains isolated standards for antimicrobial susceptibility testing; 21st infor- from diseased pigs in seven European countries to antimi- mational supplement. CLSI M100-S21. Clinical and Laboratory crobial agents licenced in veterinary medicine. Vet Microbiol Standards Institute, Wayne, Pennsylvania, USA. 2006;113:73–82. 20. Gottschalk M, Lacouture S, Bonifait L, Roy D, Fittipaldi N, 12. Zhang C, Ning Y, Zhang Z, Song L, Qiu H, Gao H. In vitro anti- Grenier D. Characterization of Streptococcus suis isolates recov- microbial susceptibility of Streptococcus suis strains isolated from ered between 2008 and 2011 from diseased pigs in Québec, clinically healthy sows in China. Vet Microbiol 2008;131:386–392. Canada. Vet Microbiol 2013;162:819–825. 13. Aarestrup FM, Jorsal SE, Jensen NE. Serological characterization 21. Kerdsin A, Dejsirilert S, Sawanpanyalert P, et al. Sepsis and spon- and antimicrobial susceptibility of Streptococcus suis isolates taneous bacterial peritonitis in Thailand. Lancet 2011;378:960. from diagnostic samples in Denmark during 1995 and 1996. Vet 22. Reams RY, Glickman LTL, Harrington DD, Bowersock TL, Microbiol 1998;60:59–66. Thacker HL. Streptococcus suis infection in swine: A retrospective 14. Aarestrup FM, Rasmussen SR, Artursson K, Jensen NE. Trends in study of 256 cases. Part 1. Epidemiologic factors and antibiotic the resistance to antimicrobial agents of Streptococcus suis isolates susceptibility patterns. J Vet Diagn Invest 1993;5:363–367. from Denmark and Sweden. Vet Microbiol 1998;63:71–80. 23. Kataoka Y, Yoshida T, Sawada T. A 10-year survey of antimi- 15. Salvarani FM, Pinto FF, Lobato FCF, et al. Concentração ini- crobial susceptibility of Streptococcus suis isolates from swine in bitória mínima (CIM) de oito antimicrobianos frente isolados Japan. J Vet Med Sci 2000;62:1053–1057. de Streptococcus suis. Braz J Vet Res Anim Sci 2008;45:41–47. 16. Higgins R, Gottschalk M. An update on Streptococcus suis iden- tification. J Vet Diagn Invest 1990;2:249–252.

2000;64:0–00 The Canadian Journal of Veterinary Research 149 FOR PERSONAL USE ONLY Short Communication Communication brève

Evaluation of a real-time polymerase chain reaction assay of the outer membrane protein P2 gene for the detection of Haemophilus parasuis in clinical samples Rebeccah McDowall, Durda Slavic, Janet I. MacInnes, Hugh Y. Cai

Abstract A real-time polymerase chain reaction (PCR) assay of the outer membrane protein (OMP) P2 gene was developed and used to test 97 putative Haemophilus parasuis pure cultures and 175 clinical tissue samples. With standard culture isolation as the gold standard, the diagnostic sensitivity and specificity of the PCR assay were determined to be 83% and 80%, respectively.

Résumé Une épreuve de réaction d’amplification en chaine par la polymérase (PCR) en temps réel pour détecter le gène P2 de la protéine de la membrane externe (OMP) fut développé et utilisé pour tester 97 cultures pures réputées d’Haemophilus parasuis et 175 échantillons de tissus. La méthode de culture standard étant considérée la méthode étalon, la sensibilité et spécificité diagnostique de l’épreuve PCR ont été déterminées comme étant respectivement 83 % et 80 %. (Traduit par Docteur Serge Messier)

Haemophilus parasuis is an important pathogen in the swine indus- had a reported sensitivity of 0.83 to 9.5 CFU per reaction and no try, causing Glässer’s disease; a syndrome that is characterized by false-positive results for A. indolicus. Recently Li et al (7) reported fibrinous polyserositis, meningitis, and arthritis (1). Historically that 2 closely related variants of the outer membrane protein (OMP) Glässer’s disease was seen sporadically in immunocompromised P2 gene are highly conserved in H. parasuis. Therefore, the objective or stressed piglets; however, in recent years H. parasuis has been of this study was to determine if it was possible to develop a real- frequently found to cause disease in pigs of all ages (2). In most time PCR (RT-PCR) assay for H. parasuis based on the OMP P2 gene diagnostic laboratories H. parasuis is identified by culture and that would have better sensitivity and specificity than the currently biochemical testing. Isolation of H. parasuis from clinical samples available tests. can be difficult owing to the bacterium’s fastidious nature and the A total of 97 isolates identified as H. parasuis by phenotypic assay fact that it is easily overgrown by other bacterial contaminants, and were tested to validate the ompP2 RT-PCR assay. This collection phenotypic identification of H. parasuis may be problematic because included 15 reference strains (serotype 1 to 15) as well as 83 sero- phenotypic variation can lead to misclassification (3). Over the past typeable and unserotypeable field isolates from swine lung and several years a number of polymerase chain reaction (PCR) assays tonsil tissues that had been isolated and characterized by standard have been described for H. parasuis detection. There are, however, phenotypic methods (1). For routine propagation, isolates were grown shortcomings in specificity and sensitivity with all of these tests, on brain–heart infusion medium containing 0.02% nicotinamide and few of them have been validated with a large number of clinical adenine dinucleotide (BHI-N). Specificity was evaluated with a panel samples. The gel-based PCR test of Oliveira, Galina, and Pijoan (4), of 21 “non-H. parasuis” bacteria typically found in swine, including based on the 16S ribosomal RNA (rRNA), had a reported sensitiv- A. indolicus (DNA extract only), A. minor, A. pleuropneumoniae, A. por- ity of 100 colony-forming units (CFU) per milliliter of H. parasuis cinus, A. suis, Bordetella bronchiseptica, Clostridium perfringens, Pantoea organisms but false-positive results for Actinobacillus indolicus. This agglomerans, Enterobacter cloacae, Erysipelothrix rhusiopathiae, Escherichia PCR test was later found to be not as sensitive as culture in animal- coli, Klebsiella oxytoca, K. pneumoniae, Mannheimia glucosida, M. haemo- challenge experiments (5). Turni and Blackall (5) described a gel- lytica, M. varigena, Mycoplasma hyopneumoniae, Pasteurella mairii (DNA based PCR assay targeting the 16S rRNA gene that had a sensitivity extract only), P. multocida, Pseudomonas aeruginosa, Proteus vulgaris, of 5 CFU per PCR reaction; however, when the assay was validated Salmonella Enteritidis phage type 13, Salmonella Heidelberg phage with 19 samples from 5 positive pigs, 6 of the samples were posi- type 7, Salmonella Typhimurium, Serratia marcescens, Staphylococcus tive only when the template was diluted. In a routine diagnostic aureus, Streptococcus suis, and Trueperella pyogenes. In addition, the per- laboratory this can be problematic and create extra work, as the formance of the ompP2 RT-PCR assay and standard culture isolation template may not be diluted routinely. Turni, Pyke, and Blackall (6) methods was compared with the use of 134 pig-lung and 41 pig-tonsil later described a real-time PCR assay targeting the infB gene that samples from clinical submissions.

Animal Health Laboratory, Laboratory Services Division, University of Guelph, 419 Gordon Street, Guelph, Ontario N1G 2W1 (McDowall, Slavic, Cai); Department of Pathobiology, University of Guelph, 419 Gordon Street, Guelph, Ontario N1G 2W1 (MacInnes). Address all correspondence to Dr. Hugh Y. Cai; telephone: 519-824-4120, ext. 54316; fax: 519-821-8072; e-mail: [email protected] Received January 17, 2013. Accepted July 4, 2013.

150 The Canadian Journal of Veterinary Research 2014;78:150–152 FOR PERSONAL USE ONLY

Extraction of DNA from the colonies grown on BHI-N or 25 mg of Table I. Results of testing 97 phenotypically identified tissue samples was done either robotically, with use of a MagNA Pure isolates of Haemophilus parasuis by real-time polymerase LC instrument and a DNA Isolation Kit I (Roche Diagnostics, Laval, chain reaction (RT-PCR) assay of the outer membrane protein Quebec), or manually by means of a DNeasy Blood & Tissue Kit P2 gene and results of 16S ribosomal RNA gene sequence (Qiagen, Toronto, Ontario), according to the manufacturers’ instruc- analysis of the isolates that were PCR-negative tions. With either method DNA was eluted in 100 mL of elution buffer. Serotype of Number of Sequencing RT-PCR For the ompP2 RT-PCR, a 20-mL reaction volume was used with H. parasuis isolates results assay results 0.8 mM each of primer HP-OmpF (TGATGGTCAATTGCGTCT) 1 1 ND 1 and HP-OmpR (CGAGTCTCATAACGACCAAA), 0.1 mM each of 2 2 ND 1 a pair of degenerate probes, HP-OmpP2-1(59FAM-AATAATTCT 3 1 ND 1 CGTTTCGGTATTTCTATCAAACA-39TAMRA) and HP-OmpP2-2 4 10 ND 1 (59FAM-AATAGTTCTCGTTTCGGTATTTCTATCAAACA- 4 2 Non-H. parasuis 2 39TAMRA), and 2 mL of DNA template. The ompP2 RT-PCR 4 3 H. parasuis 2 assay was done with use of the LC 480 Probe Master Kit (Roche 4 1 H. parasuis 1 Diagnostics) on the LC 480 instrument with the following run con- 5 9 ND 1 ditions: 95°C for 10 min and then 45 cycles of 95°C for 10 s, 57°C 5 1 H. parasuis 1 for 30 s, and 40°C for 30 s. A sample was considered positive when 6 1 ND 1 the crossing-point value was less than 40 and if there was a clear 7 1 ND 1 amplification curve. The primers and probes were designed accord- 8 1 ND 1 ing to the alignment of the DNA sequences of ompP2 from different 9 1 ND 1 H. parasuis strains available in GenBank [US National Center for 10 1 ND 1 Biotechnology Information (NCBI), Bethesda, Maryland, USA] 11 1 ND 1 with the use of Vector NTI software (Invitrogen Life Technologies, 12 1 ND 1 Burlington, Ontario) and conserved regions identified. LightCycler 13 7 ND 1 Probe Design Software 2.0 (Roche Diagnostics) was then used to 13 1 H. parasuis 2 identify possible primer/probe combinations. Owing to the genetic 14 10 ND 1 diversity of H. parasuis, the most suitable probes bound in an area 15 1 ND 1 with a single base-pair mismatch. Therefore, a pair of degenerate 5 & 12a 2 ND 1 probes was used in the PCR reactions to ensure that all strains of 5 & 12a 1 H. parasuis 2 H. parasuis could be detected by this assay. The primers and probes 5 & 13b 1 ND 1 were further analyzed by means of a BlastN (NCBI) search against Untypeable 30 ND 1 the DNA sequences deposited in the nonredundant GenBank data- Untypeable 4 Non-H. parasuis 2 base to ensure specificity. Untypeable 1 H. parasuis 1 For analytic sensitivity testing, an H. parasuis isolate (JM-486) was Untypeable 2 H. parasuis 2 grown in 5 mL of BHI-N to a concentration of 4.85 3 108 CFU/mL a The isolates cross-reacted with antiserum to serotypes 5 and 12. as confirmed by viable counts on BHI-N. Tenfold serial dilutions b The isolate cross-reacted with antiserum to serotypes 5 and 13. of the broth culture were made, DNA was extracted from 200-mL ND — Sequencing was not done. aliquots, and the pure-culture detection limit of the assay was tested in triplicate. In addition, the same dilution series of JM-486 was used to spike samples of homogenized lung tissue previously identified When pure culture and spiked tissue samples were tested with as not being infected with H. parasuis by the conventional PCR the ompP2 RT-PCR assay, 2 and 68 CFU per reaction were detected, described by Angen et al (8), and the detection limit of H. parasuis respectively. When the panel of non-H. parasuis organisms were in those samples was tested in triplicate. tested with the ompP2 RT-PCR assay, no false-positive results were To verify the identity of all H. parasuis isolates that were PCR- obtained. negative for ompP2 we conducted 16S rRNA gene sequence analy- The diagnostic sensitivity and specificity of the assay were deter- sis (9). Some isolates of type 4 and 5 and untypeable were also mined with 175 clinical samples (134 of lung and 41 of tonsil), sequenced as most of the misidentified isolates belonged to these H. parasuis having been isolated from 23. With the ompP2 RT-PCR serotypes (see below). assay, of the 23 culture-positive samples 19 (83%) were PCR-positive, All 15 H. parasuis reference strains were correctly identified by and of the 152 culture-negative samples 121 (80%) were PCR- the ompP2 RT-PCR assay. Of the 97 isolates originally identified negative. Therefore, the ompP2 RT-PCR assay has a diagnostic phenotypically as H. parasuis, 6 were confirmed by 16S rRNA gene sensitivity of 83% and a specificity of 80%. The lower calculated sequencing as non-H. parasuis (Table I). Among the true H. parasuis specificity may be due to the lower sensitivity of culture. isolates, 92% (84/91) were positive by the ompP2 RT-PCR assay; the In the past decade several PCR assays have been described for the misidentified H. parasuis isolates belonged to serotype 4 (3 strains) detection of H. parasuis in swabs or tissue samples, however, with or serotype 13 (1 strain), were untypeable (2 strains), or cross-reacted relatively low diagnostic sensitivity or specificity. The specificity of with antiserum to serotypes 5 and 12 (1 strain). the ompP2 RT-PCR assay developed in this study was determined

2000;64:0–00 The Canadian Journal of Veterinary Research 151 FOR PERSONAL USE ONLY to be 80% with culture used as the gold standard, a figure similar References to that obtained with the PCR assays previously described. Turni et al (6) described a real-time PCR assay that detected H. parasuis in 1. Quinn PJ, Markey BK, Leonard FC, Fitzpatrick ES, Fanning S, 66% of swabs from tissue samples collected from pigs challenged Hartigan PJ. Veterinary microbiology and microbial disease. with H. parasuis, whereas culture isolation detected H. parasuis in Second Edition. ed. Chichester, West Sussex, UK: Wiley-Blackwell; only 58% of the swabs. Similarly, the gel-based PCR test of Oliveira 2011. et al (4) was positive for 26 of 30 samples, whereas culture isolation 2. MacInnes JI, Gottschalk M, Lone AG, et al. Prevalence of detected H. parasuis in only 18 of the 30 samples. We conclude that Actinobacillus pleuropneumoniae, Actinobacillus suis, Haemophilus when the ompP2 RT-PCR assay described here was validated with parasuis, Pasteurella multocida, and Streptococcus suis in representa- field samples, it performed similarly to previously described PCRs tive Ontario swine herds. Can J Vet Res. 2008;72:242–248. and therefore can be used as an additional diagnostic tool. 3. Kielstein P, Rapp-Gabrielson VJ. Designation of 15 serovars of Haemophilus parasuis on the basis of immunodiffusion using Acknowledgments heat-stable antigen extracts. Journal of clinical microbiology. 1992;30:862–865. This project was supported by the Ontario Ministry of Agriculture, 4. Oliveira S, Galina L, Pijoan C. Development of a PCR test to Food and Rural Affairs Agreement with the University of Guelph diagnose Haemophilus parasuis infections. J Vet Diagn Invest. through the Animal Health Strategic Investment Fund managed 2001;13:495–501. by the Animal Health Laboratory of the University of Guelph. 5. Turni C, Blackall P. Comparison of sampling sites and detection Dr. Oystein Angen kindly provided the DNA extracts of P. mairii methods for Haemophilus parasuis. Aust Vet J. 2007;85:177–184. and A. indolicus. 6. Turni C, Pyke M, Blackall PJ. Validation of a real-time PCR for Haemophilus parasuis. J Appl Microbiol. 2010;108:1323–1331. Conflict of interest 7. Li P, Bai J, Li JX, et al. Molecular cloning, sequencing, and expres- All authors have no financial and personal relationships with sion of the outer membrane protein P2 gene of Haemophilus other people or organizations that could inappropriately influence parasuis. Res Vet Sci. 2012;93:736–742. (bias) their work. 8. Angen O, Oliveira S, Ahrens P, Svensmark B, Leser TD. Development of an improved species specific PCR test for detec- Role of the funding source tion of Haemophilus parasuis. Vet Microbiol. 2007;119:266–276. The study sponsors had no involvement in the study design, in 9. Cai H, Archambault M, Prescott JF. 16S ribosomal RNA sequence- the collection, analysis and interpretation of data, in the writing of based identification of veterinary clinical bacteria. J Vet Diagn the manuscript, and in the decision to submit the manuscript for Invest. 2003;15:465–469. publication.

152 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY Short Communication Communication brève

Effect of native Lactobacillus murinus LbP2 administration on total fecal IgA in healthy dogs Luis Delucchi, Martín Fraga, Karen Perelmuter, Claudia Della Cella, Pablo Zunino

Abstract The objective of the present work was to determine the effect of Lactobacillus murinus strain LbP2 on canine fecal immunoglobulin A (IgA) levels. Seven dogs were orally treated with a 3-mL suspension of L. murinus LbP2 containing 5 3 109 colony-forming units on alternate days for 2 wk. Six dogs were treated with 3 mL of phosphate-buffered saline as placebo. Fecal samples were taken from the rectal ampulla on days 0 and 16, and the total canine fecal IgA concentration was determined with an immunoperoxidase assay kit. The IgA levels of individual dogs were compared with the nonparametric Wilcoxon test. Differences were considered significant when the P-value was less than 0.05. An increase in the total fecal IgA concentration was observed in the 7 dogs after treatment with L. murinus LbP2 (P = 0.01796). No differences were detected between the initial total fecal IgA values and those obtained at the end of placebo treatment. Thus, after oral administration L. murinus LbP2 showed potential immunomodulatory effects, an important property to assess in a microorganism being considered for use as a probiotic.

Résumé L’objectif de la présente étude était de déterminer l’effet de la souche Lactobacillus murinus LbP2 sur les niveaux d’immunoglobulines A (IgA) fécales chez le chien. Sept chiens ont reçu oralement 3 mL d’une suspension de L. murinus LbP2 contenant 5 3 109 unités formatrices de colonies à chaque 2 jours pendant 2 semaines. Six chiens ont reçu 3 mL de saline tamponnée comme placebo. Des échantillons de fèces ont été prélevés au niveau de l’ampoule rectale aux jours 0 et 16, et la concentration d’IgA canine totale déterminée au moyen d’une épreuve d’immunoperoxydase. Les niveaux d’IgA des chiens individuels ont été comparés avec le test non-paramétrique de Wilcoxon. Les différences étaient considérées significatives lorsque la valeur de P était inférieure à 0,05. Une augmentation dans la concentration d’IgA fécale totale a été observée dans les sept chiens après traitement avec L. murinus LbP2 (P = 0,01796). Aucune différence ne fut détectée entre les valeurs initiales des IgA fécales totales et celles à la fin du traitement avec le placebo. Ainsi, l’administration orale de L. murinus LbP2 montre des effets immunomodulateurs potentiels, une propriété intéressante à évaluer chez un microorganisme considéré pour usage comme probiotique. (Traduit par Docteur Serge Messier)

Probiotics are defined as “live microorganisms which when One of the dominant lactobacilli in the dog’s intestinal microbiota administered in adequate amounts confer a health benefit on the and milk, L. murinus, plays an important role in the immunity of host” (1). The beneficial effects and the use of probiotics in the man- puppies (6,7). Different L. murinus strains have, in vitro, shown agement of specific diseases are well-accepted (2). Commonly used interesting properties for probiotic use, exhibiting acid and bile salt bacterial probiotics include Lactobacillus spp. and Bifidobacterium resistance, adhesion to canine intestinal mucus, and antimicrobial spp.; Lactococcus lactis and some Enterococcus species have also been activity against specific bacterial pathogens (8,9). An in-vivo assay used (3). in dogs proved the enteric persistence and safety of L. murinus (10). The mechanisms of action of probiotics differ and are not unique. The aim of the present work was to evaluate the immunomodula- One potential mechanism is based on the ability to modulate the tion potential of L. murinus strain LbP2 through the effect on total mucosal immune system and the local production of IgA. In some fecal IgA production in orally treated dogs. cases probiotics have been associated with total and pathogen- One group of 7 and another of 6 healthy dogs between 3 and specific IgA levels upon infection while typically not inducing the 7 y old from the Veterinary Faculty of the University of Uruguay, production of specific antiprobiotic IgA. Treatment with L. casei Montevideo, were used; all animal procedures were approved by the induced a significant increase in the numbers of IgA-producing cells Bioethics Committee of the Veterinary Faculty. The probiotic-treated in the small-bowel lamina propria of mice (4). However, not all pro- group included 5 females (3 that were 7 y old and 1 each that were biotic strains are equal in terms of their effects on IgA production (5). 4 and 5 y old) and 2 males (7 and 3 y old). The group treated with The domestic dog (Canis lupus familiaris) plays an important role in phosphate-buffered saline (PBS) included 4 females (3 that were 7 y our society and has been used as a model to study various diseases. old and 1 that was 5 y old) and 2 males (3 and 5 y old). The animals

Departamento de Patología y Clínica de Pequeños Animales, Facultad de Veterinaria, Universidad de la República, Lasplaces 1550, CP 11600, Montevideo, Uruguay (Delucchi, Della Cella); Instituto de Investigaciones Biológicas Clemente Estable, Avenida Italia 3318, CP 11600, Montevideo, Uruguay (Fraga, Perelmuter, Zunino). Address all correspondence to Dr. Luis Delucchi; telephone/fax: 598-26287775; e-mail: [email protected] Received May 17, 2013. Accepted July 11, 2013.

2014;78:153–155 The Canadian Journal of Veterinary Research 153 FOR PERSONAL USE ONLY

3 Table I. Changes in total fecal immunoglobulin A (IgA) 2.7 concentration in 7 dogs treated with Lactobacillus murinus strain LbP2 and 6 dogs treated with phosphate-buffered 2.4 saline as placebo for 2 wk 2.1 Fecal IgA level, mg per gram of feces; treatment 1.8 Lactobacillus murinus Phosphate-buffered 1.5 strain LbP2 saline 1.2 Day 0 Day 16 Day 0 Day 16 0.366 0.581 0.06 0.101 Fecal IgA mg/g 0.9 0.127 0.584 0.75 0.32 0.6 0.038 0.096 0.49 0.08 0.3 0.385 0.821 2.77 0.05 0.269 1.04 0.05 0.055 0 A B C D 0.037 1.01 0.81 0.81 1.36 2.621 Figure 1. Total fecal IgA concentration in dogs before and after treat- P = 0.01796 P . 0.05 ment with Lactobacillus murinus strain LbP2 (A, B) or phosphate-buffered saline (C, D). Each plot shows the median (line within box), 25th and 75th percentiles (box) and extremes. The difference in median concen- tration after treatment with L. murinus was significant (P = 0.01796), whereas the difference after placebo treatment was not (P . 0.05). cate measures were obtained. The results were expressed as mil- ligrams of IgA per gram of feces. were fed once a day with 270 g of dry commercial food (Royal Canin, A 4-parameter logistics curve was constructed from the results Buenos Aires, Argentina) from 10 d before the start of the study. All observed for the standards to interpolate the values for the test the dogs were kept at the same place and isolated. At the beginning samples. The IgA levels of individual dogs were compared with the and at the end of the study (days 0 and 16) their body condition, nonparametric Wilcoxon test. Differences were considered significant body weight, and fecal consistency were determined. when the P-value was less than 0.05. The probiotic-treated group received orally 3 mL of a suspen- No significant changes in body condition, body weight, or fecal sion of L. murinus LbP2 containing 5 3 109 colony-forming units on consistency were observed in the dog populations before and after alternate days for 2 wk, for a total of 8 doses. The bacteria had been treatment (data not shown). The standard curve constructed with the grown in MRS broth for 48 h, washed twice, and resuspended in PBS use of the standard absorbance values showed an r2 value of 0.989. at an adequate concentration. The control group received orally 3 mL The total fecal IgA concentration increased significantly (P = 0.01796) of PBS on the same days as the probiotic-treated group. in the 7 probiotic-treated dogs: immediately before treatment the Fecal samples were taken from the rectal ampulla of each dog on values ranged from 0.037 to 1.36 (median 0.269) mg/g, and at the study days 0 and 16 (2 d after the last dose) and suspended in PBS end of treatment they ranged from 0.096 to 2.621 (median 0.821) (w/v ratio 1:4) supplemented with ethylene diamine tetraacetic acid mg/g (Figure 1, Table I). The total fecal IgA concentration did not and phenylmethylsulfonyl fluoride, 0.5 mM each, to inhibit protease change significantly (P . 0.05) in the 6 PBS-treated dogs: immedi- activity (11). The samples were homogenized by means of a vortex ately before treatment the values ranged from 0.05 to 2.77 (median blender at maximum speed for 30 min at 4°C and the homogenates 1.53) mg/g, and at the end of treatment they ranged from 0.05 to 0.81 centrifuged twice at 10 000 3 g for 20 min at 4°C. The supernatants (median 0.2367) mg/g. When the total fecal IgA levels of the 2 groups were collected and stored at −80°C for subsequent analysis. of dogs were compared, no significant differences were detected at An immunoperoxidase assay kit (Immunology Consultants the beginning of the study (P . 0.05), whereas there were significant Laboratory, Portland, Oregon, USA) was used to determine the differences at the end of the study (P = 0.02681). total canine fecal IgA concentration. The standard was prepared in From the results of the present study it can be concluded that 7 serial dilutions of 0 to 1000 ng/mL. Successively, 100-μL aliquots oral administration of L. murinus LbP2 significantly influences the of the fecal samples diluted 1:500 were added to predesignated wells production of total canine fecal IgA. Not all probiotics are equal coated with antibody against dog IgA provided by the manufacturer. in terms of their effects on IgA production. In a study comparing The microplates were incubated for 30 min and then aspirated and different probiotic properties, including intestinal IgA production, washed 3 times. Antibody conjugated with horseradish peroxidase Saccharomyces boulardii induced greater production of IgA than did was added, and the plates were incubated for another 30 min and Escherichia coli, B. animalis, and L. casei, whereas B. animalis and then washed and blotted again. Next, 100 μL of cromogen-substrate L. casei were better as pathogen inhibitors than S. boulardii (11). Thus, solution (3,39,5,59-tetramethybenzidine and hydrogen peroxide in a particular property such as immunomodulation through local IgA citric acid buffer, pH 3.3) was added to each well, and the plates production is not necessarily associated with other probiotic-related were incubated for 10 min. Finally, stop solution (0.3 M sulfuric acid) attributes. No effect on fecal IgA concentration was observed after was added to end the reaction, and the absorbance was determined administration of the probiotic Enterococcus faecium SF68 to treat (at 450 nm) with a Varioskan Flash plate reader (Thermo Scientific, giardiasis in dogs (13). Others have reported that the use of pre- Asheville, North Carolina, USA). For standards and samples, dupli- biotics such as inulin in association with probiotics could have a

154 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

synergistic effect on the modulation of gut immunity, including an rhamnosus and Bifidobacterium lactis modulates intestinal immune increase in IgA production (5). All these findings suggest that there functions in rats. J Nutr 2004;134:153–156. is a remarkable diversity of stimuli that can lead to an enhanced 6. Rinkinen ML, Koort JM, Ouwehand AC, Westermarck E, enteric immune response. In addition, canine fecal IgA levels are Björkroth KJ. Streptococcus alactolyticus is the dominating cul- highly variable, and the influence of diverse factors such as age and turable lactic acid bacterium species in canine jejunum and feces breed has been reported (14,15). of four fistulated dogs. FEMS Microbiol Lett 2004;230:35–39. As in the previous in-vivo assay in which L. murinus LbP2 was 7. Martin R, Olivares M, Pérez M, et al. Identification and evalua- administered to dogs (10), the microorganism did not affect body tion of the probiotic potential of lactobacilli isolated from canine condition or fecal consistency. This could indicate that no inflamma- milk. Vet J 2010;185:193–198. tory disorders were induced by the probiotic. It is important to take 8. Nardi R, Santoro M, Oliveira J, et al. Purification and molecu- into account that L. murinus LbP2 treatment did not significantly lar characterization of antibacterial compounds produced by affect the dogs fecal microbiota as observed in a previous study (10) Lactobacillus murinus strain L1. J Appl Microbiol 2005;99:649–656. or the current one. 9. Perelmuter K, Fraga M, Zunino P. In vitro activity of potential Considering these encouraging results, further studies are being probiotic Lactobacillus murinus isolated from the dog. J Appl conducted to assess other potential probiotic effects of L. murinus Microbiol 2008;104:1718–1725. LbP2 in dogs. 10. Perelmuter K, Fraga M, Delucchi L, Zunino P. Safety assessment and enteric colonization ability of a native canine Lactobacillus References murinus strain. World J Microbiol Biotechnol 2011;27:1725–1730. 11. Sainz T, Perez J, Fresan MC, et al. Histological alterations and 1. Guidelines for the Evaluation of Probiotics in Food. Report immune response induced by pet toxin during colonization of a Joint FAO/WHO Working Group on Drafting Guidelines with enteroaggregative Escherichia coli (EAEC) in a mouse model for the Evaluation of Probiotics in Food. 2002. Available from: infection. J Microbiol 2002;40:91–97. http://www.who.int/foodsafety/fs_management/en/probiotic_ 12. Martins FS, Silva AA, Vieira AT, et al. Comparative study of guidelines.pdf Last accessed January 4, 2014. Bifidobacterium animalis, Escherichia coli, Lactobacillus casei and 2. Narayan SS, Jalgaonkar S, Shahani S, Kulkarnin VN. Probiotics: Saccharomyces boulardii probiotic properties. Arch Microbiol 2009; Current trends in the treatment of diarrhea. Hong Kong Med J 191:623–630. 2010;16:213–218. 13. Simpson KW, Rishniw M, Bellosa M, et al. Influence of 3. De Vrese M, Schrezenmeir J. Probiotics, prebiotics, symbiotics. Enterococcus faecium SF68 probiotic on giardiasis in dogs. J Vet Adv Biochem Eng Biotechnol 2008;111:1–66. Intern Med 2009;23:476–481. 4. Galdeano CM, Perdigon G. The probiotic bacterium Lactobacillus 14. Zaine L, Ferreira C, Gomes M de O, et al. Faecal IgA concentration casei induces activation of the gut mucosal immune system is influenced by age in dog. Br J Nutr 2011;106(S1):S183–S186. through innate immunity. Clin Vaccine Immunol 2006;13: 15. Littler RM, Batt RM, Lloyd DH. Total and relative deficiency of 219–226. gut mucosal IgA in German shepherd dogs demonstrated by 5. Roller M, Rechkemmer G, Watzl B. Prebiotic inulin enriched with faecal analysis. Vet Rec 2006;158:334–341. oligofructose in combination with the probiotics Lactobacillus

2000;64:0–00 The Canadian Journal of Veterinary Research 155 FOR PERSONAL USE ONLY Short Communication Communication brève

Ex vivo evaluation of 7 polydioxanone for closure of equine ventral midline celiotomies Stacy L. Anderson, Jose L. Bracamonte, Steve Hendrick

Abstract The objective of this study was to compare the bursting strength (BS) and mode of failure (MF) of ventral midline (VM) celiotomies closed with USP 7 polydioxanone (7PD) in 1 or 2 simple continuous sections. A bursting strength model, consisting of inserting and inflating a 200-L polyurethane bladder through a 25-cm VM celiotomy, was used on 15 fresh equine cadavers. Celiotomies were closed using 7PD in 2 separate sections (4 knots), 2 continuous sections (3 knots), or a single section (2 knots) using a simple continuous pattern. The horses’ signalment, body weight, number of total knots, MF, and BS were recorded and analyzed statistically for interactions. No difference was found between the BS of VM celiotomies closure types (P = 0.4). All celiotomy/ suture constructs failed at the abdominal wall. The celiotomy closure types evaluated in this study provided a secure method of closure in VM celiotomies in vivo.

Résumé L’objectif de la présente étude était de comparer la force d’éclatement (BS) et le mode d’échec (MF) de laparotomies par la ligne ventrale médiale (VM) refermées avec du 7 polydioxanone USP (7PD) en une ou 2 sections continues. Un modèle de force d’éclatement, consistant en l’insertion et le gonflement une vessie de 200 L en polyuréthane via une laparotomie de 25 cm fut utilisé sur 15 cadavres frais de cheval. Les laparotomies étaient refermées en utilisant le 7PD en deux sections séparées (4 nœuds), 2 sections continues (3 nœuds), ou une section simple (2 nœuds) au moyen d’un patron simple continu. L’historique des chevaux, leur poids corporel, le nombre total de nœuds, MF, et BS ont été enregistrés et analysés statistiquement pour les interactions. Aucune différence ne fut trouvée entre les BS et les types de fermetures de VM des laparotomies (P = 0,4). Toutes les laparotomies/modes de suture ont lâché au niveau de la paroi abdominale. Les types de fermeture des laparotomies évalués dans ce projet ont fourni une méthode sécuritaire de fermeture par la VM de laparotomies in vivo. (Traduit par Docteur Serge Messier)

Basic surgical principles state that the suture chosen to close a We hypothesized that there would be no difference in bursting surgical wound should be as strong as the tissue through which is strength or mode of failure among equine VM celiotomies closed it placed in the immediate post-operative period (1). Though many with 7PD in 2 separate simple continuous sections (4 knots), 2 con- sizes and type of sutures have been evaluated in the closure of tinuous simple continuous sections (2 sections tied to each other in equine celiotomies (2–8), none, other than braided USP 7 polydioxa- the middle to create 3 knots), or a single simple continuous section none (7PD) and looped USP 2 lactomer 9-1 (3,8,9), have fulfilled the (2 knots). Therefore, our objective was to test the bursting strength requirements of a suture to be as strong as the equine abdominal and mode of failure of these different closure types using a previ- wall. Both 7PD and looped USP 2 lactomer 9-1 consistently resisted ously reported ex vivo equine celiotomy bursting strength model failure, because the abdominal wall failed before the suture lines, in (2,8,9). separate equine ex vivo bursting strength trials (3,8,9) likely due to Fifteen horses, euthanized for reasons unrelated to this study, were their large working diameter of 1-mm and because a simple continu- randomly placed into 1 of 3 groups as determined by the number ous pattern was used to close the celiotomies. It has previously been of knots in celiotomy closure (Figure 1). The number of horses per shown that a simple continuous pattern has the greatest bursting group was determined by using a sample size calculation based on strength compared to other patterns used to close celiotomies (10,11). previous data and a beta error of 0.80. Inclusion criteria included: age Although the authors primarily close equine celiotomies with ≥ 2 years, body weight . 300 kg and , 600 kg, no previous history 7PD in 2 simple continuous sections, celiotomies that are less than of abdominal surgery, and no obvious abdominal wall abnormalities 30 cm in length may be closed in a single simple continuous section (i.e., umbilical herniation). The age (year), gender, breed, and body using 7PD due to the long working length of the suture (1.5 m). To weight (kg) were recorded for each horse. All procedures and experi- our knowledge, there are no studies that have evaluated the bursting mental protocols were approved by the University of Saskatchewan strength of equine VM celiotomies closed with any type of suture Committee on Animal Care and Supply and the University of in a single simple continuous section compared to closure of VM Saskatchewan Animal Research Ethics Board. celiotomies by multiple continuous sections.

Western College of Veterinary Medicine, University of Saskatchewan, 52 Campus Drive, Saskatoon, Saskatchewan S7N 5B4. Address all correspondence to Dr. Stacy Anderson; telephone: (306) 371-8208; e-mail: [email protected] Received February 19, 2013. Accepted April 23, 2013.

156 The Canadian Journal of Veterinary Research 2014;78:156–160 FOR PERSONAL USE ONLY

Figure 1. Components of the experimental set-up. A — Air compressor, B — polyurethane bladder, C — personal computer with data acquisition software depicting a bursting trial, D — pressure transducer, E — digital video camera, and F — nylon strap across inguinal region.

Immediately following euthanasia with an overdose of pento- mies were closed using a single simple continuous section result- barbital, cadavers were placed in dorsal recumbency. A 40-cm skin ing in the formation of 2 knots. Five celiotomies were closed with incision overlying the VM celiotomy site was created, the skin was 2 separate simple continuous sections resulting in the creation of dissected away from the incision site, and a VM celiotomy was cre- 4 knots. The remaining 5 celiotomies were closed using 2 continu- ated beginning at the umbilical scar and extending 25-cm cranially ous simple continuous sections where the suture strands were tied through the linea alba. to each other in the middle of the celiotomy resulting in the creation Figure 1 illustrates the experimental set-up. Following celiotomy, of 3 knots. For all celiotomy closures, tissue bites were standardized stab incisions were made between the 14th and 15th rib, in the mid- using a square template 15-mm from the edge of the incision and dle one third of the left and right intercostal space, to provide access 15-mm apart. Four throws were used to secure each surgeon’s knot for ingress and egress tubing. A custom 0.9 m 3 1.2 m rectangular, (6 throws total) and suture tags were trimmed to 5-mm in length. All 200-L capacity, polyurethane bladder (New World Manufacturing, celiotomies were created and closed by one surgery resident (SLA). Cloverdale, California, USA) with integrated ingress and egress tub- The skin was left unopposed to allow unimpeded observation of ing was inserted into the abdominal cavity. The ingress tubing was the suture and body wall during testing. After celiotomy closure, a exteriorized through the right lateral stab incision and attached via 5-cm wide nylon ratchet strap was placed caudal to the celiotomy, quick couplers to an 8-L capacity air compressor (Mastercraft Air just cranial to the pelvis, and tightened to prevent herniation of the Compressor; Canadian Tire, Toronto, Ontario) with digital air flow abdominal organs through the inguinal fascia and/or rectum. regulator (Smart-Trak 50 Series Mass Flow Meters & Controller; All celiotomy/suture constructs were tested within 1 h of Sierra, Monterey, California, USA). The egress tubing was exterior- horses being euthanized. Bursting strength and mode of failure of ized through the left lateral stab incision and connected to a pres- celiotomy/­suture constructs was evaluated by inflating the intra- sure transducer (P55 Compact Differential Pressure Transducer; abdominal bladder with compressed room air until celiotomy/ Validyne Engineering, Northridge, California, USA) that was, suture construct failure. Celiotomy bursting strength was defined in turn, connected to an analog-to-digital data acquisition card as the maximum bursting pressure (mmHg) recorded prior to a sud- (National Instruments, Austin, Texas, USA) and a computer with den decrease in the slope of pressure versus time curve. Still frame data acquisition software (LabView 2009, National Instruments). capturing (Final Cut Express 4.0; Apple, Cupertino, California, USA) Once the bladder was in place, the VM celiotomies were closed was used for evaluation of the mode of failure for each celiotomy/ using 7PD (PDS II, braided polydioxanone, 1-mm diameter, 1.5-m suture construct after testing was completed. Mode of failure was length, swaged onto a 1/2 curve 65 mm taper needle; Ethicon, defined as either suture failure or abdominal wall failure. Abdominal Markham, Ontario) and 1 of the 3 methods (Figure 2). Five celioto- wall failure was further defined by location of failure. Failure

2000;64:0–00 The Canadian Journal of Veterinary Research 157 FOR PERSONAL USE ONLY

terms. The residuals were then evaluated for their normality and homoscedasticity. The median age of the horses was 10.0 6 2.0 y (range: 3 to 28 y). The mean body weight (6 standard deviation) was 457.6 6 13.3 kg. There were 7 geldings, 7 mares, and 1 stallion. Breeds represented included Quarter horse-type (n = 10), Thoroughbred (n = 4), and Arabian (n = 1). The variation of age, body weight, gender, or breed was not significantly different among celiotomy closure groups. The post hoc power calculation to detect a difference in burst- ing strength among groups using 5 horses per group was 0.61. The mean abdominal bursting strength for all constructs was 404.6 6 19.0 mmHg. There was no significant difference among the bursting strengths of the celiotomy closure types (P = 0.4, Table I). Age had a significant negative effect on bursting strength where bursting strength decreased 6.2 mmHg for every year of increase in horses’ age (P = 0.03; 95% CI: 211.6 to 20.8). All celiotomies failed at the abdominal wall while the suture lines remained intact (Table I). There was no association between the celiotomy closure type and location of failure (P = 0.8). In this study, we compared the bursting strength and mode of failure among VM celiotomies closed with 7PD using a single simple continuous section, 2 continuous simple continuous sections, and 2 separate simple continuous sections. Our results show that there was no difference in the bursting strength, nor the mode of failure, among celiotomy closure types. However, increasing age had a nega- tive effect on celiotomy bursting strength regardless of closure type. In addition, our study was slightly underpowered (0.61) to detect Figure 2. Illustration of celiotomy closure types using 7PD. A — Single a difference in bursting strength among groups despite the results simple continuous section, B — 2 separate simple continuous sections, of a pre-study sample size calculation indicating that 3 horses were and C — 2 continuous simple continuous sections. necessary per group to achieve a beta error of 0.80. We found that 7PD used in a simple continuous pattern to close directly along the suture line was termed “adjacent to the celiotomy;” equine VM celiotomies, regardless of the number of knots within while failure at any distance away from the celiotomy was termed the suture line, resisted failure (abdominal wall failed before the “remote from the celiotomy.” Bursting strength and mode of failure suture line). This finding is consistent with results from previous were recorded for each celiotomy bursting trial. All biomechanical studies done using the same model to evaluate 7PD (8,9). It has bursting trials were digitally recorded. been shown in many studies that the weakest part of a suture line Data were analyzed using a commercial software package is the knot due to the shear stress that arises at the knot/suture line (Stata 12; StataCorp, College Station, Texas, USA). For all compari- interface when distraction forces are applied (2,4–8). In fact, a knot sons, a P , 0.05 was considered significant. Descriptive statistics and within a line of suture may decrease the strength of the suture line by a post-hoc power calculation were performed. A Shapiro-Wilk test of 30% to 35% (11). Celiotomies closed in a simple continuous pattern normality was done on all variables to determine distribution. An resist failure and have greater bursting strengths compared to inter- analysis of variance (ANOVA) was performed to test for difference rupted patterns, possibly because of fewer knots in the suture line in body weight among celiotomy closure groups. A Kruskall-Wallis (2,10,11). When USP 3 polyglactin 910 and USP 2 polyglactin were analysis of variance was performed to test for difference in age evaluated in a similar ex vivo bursting strength model using a simple among celiotomy closure groups. A Fisher’s exact test was performed continuous pattern, 71% and 60% of celiotomies failed at the suture, to test for differences in gender and breed among celiotomy closure respectively (2,8). In those 2 studies, the primary location of suture groups. A Fisher’s exact test was used to assess difference in location failure was at a knot(s) within a simple continuous pattern. In this of failure (remote or adjacent to the celiotomy) by celiotomy closure study, 7PD resisted failure at the knot within a simple continuous type. Linear regression analysis with manual backward elimination section of suture even when potentially supraphysiologic pressures was used to test the relationship between bursting strength and were applied in a single cycle to the celiotomy/suture constructs. independent variables (age, breed, gender, body weight, celiotomy Based on the results of this study, there is no greater risk from closure type). Celiotomy closure type was forced into all models. 2 simple continuous sections of 7PD failing if the 2 sections are Variables were removed sequentially when they were the least knotted in the middle. By creating only 3 knots, the suture volume significant within the model and interactions between significant is decreased because the greatest suture volume in a line of suture variables were tested. The linearity of continuous independent vari- is contained within a knot (12). By decreasing suture volume, tissue ables (age) was evaluated by testing the significance of polynomial reaction would also theoretically be decreased leading to improved

158 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY

Table I. Mean (standard deviation) of bursting strength; mode of failure for each group. Single simple continuous section (2K), 2 continuous simple continuous sections (3K), and 2 separate simple continuous sections (4K) 2K 3K 4K Bursting strength (mmHg) 6 standard deviation 387.5 6 44.4 393.8 6 18.1 432.5 6 34.2 Failure of abdominal wall Adjacent to celiotomy 4 3 2 Remote from celiotomy 1 2 3 Failure of suture 0 0 0 Total 5 5 5

wound healing (12). The question remains whether the larger size mended as the suture material of choice to close VM celiotomies of 7PD would cause a greater tissue reaction compared with other in horses. sutures commonly used to closed equine VM celiotomies (e.g., USP 2 or 3 polyglactin 910) in vivo that may compromise wound healing. References Further research is warranted to evaluate tissue reaction and com- plications of incisional healing associated with the clinical use of 1. Van Winkle W, Hasting JC. Considerations in the choice of suture 7PD for closure of equine VM celiotomies. material for various tissues. Surg Gynecol Obstet 1972;135: Multiple studies have shown that increased surgery time during 113–116. equine exploratory celiotomy leads to increased post-operative 2. Magee AA, Galuppo LD. Comparison of incisional bursting morbidity (13,14). By having flexibility in the number of simple strength of simple continuous and inverted cruciate suture pat- continuous sections required to securely close a celiotomy, a surgeon terns in the equine linea alba. Vet Surg 1999;28:442–447. may reduce valuable surgery time. Unfortunately, celiotomy closures 3. Hassan KA, Galuppo LD, Van Hoogmoed LM. An in vitro com- were not timed in this study as all celiotomies were closed by one parison of two suture intervals using braided absorbable loop person; but, it seems obvious that if 2 surgeons are available, as is suture in the equine linea alba. Vet Surg 2006;35:310–314. the case at most academic institutions, celiotomy closure time can 4. Trostle SS, Wilson DG, Stone WC, et al. A study of the biome- be decreased by closing celiotomies in 2 sections with 2 surgeons chanical properties of the adult equine linea alba: Relationship working simultaneously. of tissue bite size and suture material to breaking strength. Vet Similar to other studies using the same ex vivo model, increasing Surg 1994;23:435–441. age was found to negatively affect celiotomy bursting strength. 5. Fierheller EE, Wilson DG. An in vitro biomechanical com- In humans, age-related muscle degeneration is a well-recognized parison of the breaking strength and stiffness of polydiaxanone phenomenon (15). Age-related muscle degeneration has not been (sizes 2,7) and polyglactin 910 (sizes 3,6) in the equine linea alba. recognized in horses, but ventral abdominal failure with or without Vet Surg 2005;34:18–23. prepubic tendon rupture is empirically recognized more frequently 6. Mulon PY, Zhim F, Yahia LH, et al. The effect of six knotting in the older, albeit often multiparous, broodmare suggesting that methods on the biomechanical properties of three large diameter perhaps age-related muscle degeneration might occur in horses (16). absorbable suture materials. Vet Surg 2010;39:561–565. Further research may be warranted to elucidate changes in muscle 7. Campbell EJ, Bailey JV. Mechanical properties of suture materi- strength and integrity as horses age. als in vitro and after in vivo implantation in horses. Vet Surg The authors recognize that it would be necessary to know the 1992;21:355–361. bursting strength of the equine ventral abdominal wall in order 8. Anderson SL, Bracamonte JL, Hendrick S, Carmalt JL, Wilson DG. to make direct comparisons between the bursting strength of the Ex vivo comparison of USP 7 polydioxanone and USP 2 polygla- equine abdominal wall and celiotomies closed with 1 or 2 simple ctin 910 for closure of ventral midline celiotomies in horses. Vet continuous sections of 7PD. Even so, by not failing in multiple ex vivo Surg 2013;42:463–7. studies, the authors propose that 7PD is an appropriate choice of 9. Anderson SL, Bracamonte JL, Hendrick S, Carmalt JL, Wilson DG. suture for closing equine celiotomies in 1 or 2 simple continuous Ex vivo comparison of bursting strength of ventral midline sections if surgeons abide by the basic surgical principle that the and right ventral paramedian celiotomies in horses. Vet Surg suture chosen to close a tissue should be as strong as the tissue itself 2013;42:468–72. in the immediate post-operative period (1), although the results of 10. Poole GV, Meredith JW, Kon ND, et al. Suture technique and this study can only conclude the strength of 7PD is sufficient in the wound-bursting strength. Am Surg 1984;50:569–572. immediate post-operative period to support the equine linea alba. 11. Seid MH, MacDaniels-Owens M, Poole GV, et al. A randomized Further investigation is warranted to assess tissue interactions with trial of abdominal incision suture technique and wound strength 7PD that might affect wound healing in vivo before it can be recom- in rats. Arch Surg 1995;130:394–397.

2000;64:0–00 The Canadian Journal of Veterinary Research 159 FOR PERSONAL USE ONLY

12. Van Rijssel EJC, Brand R, Admiraal MD, et al. Tissue reaction 15. Degens H. Age-related skeletal muscle dysfunction: Causes and and surgical knots: The effect of suture size, knot configuration, mechanisms. J Musculoskelet Neuronal Interact 2007;7:246–252. and knot volume. Obstet Gynecol 1989;74:64–68. 16. Kummer MR, Stick JA. Abdominal hernia. In: Auer JA, Stick JA, 13. Proudman CJ, Edwrads GB, Barnes J, French NP. Factors affec- eds. Equine Surgery. 4th ed. St. Louis, Missouri: Saunders Elsevier, tion long-term survival of horses recovering from surgery of the 2012:512. small intestine. Eq Vet J 2005;37:360–365. 14. Roussel AJ, Cohen ND, Hooper RN, Rakestraw PC. Risk factors associated with development of postoperative ileus in horses. J Am Vet Med Assoc 2001;219:72–78.

160 The Canadian Journal of Veterinary Research 2000;64:0–00 FOR PERSONAL USE ONLY Produits et services Supplies and services

Reaching Canada’s Veterinarians Canadian Journal Get your message into of Veterinary Research available online now! The Canadian Veterinary Journal Articles are now easily accessible For more information contact: over the Internet at Laima Laffitte Advertising Manager Tel.: (613) 673-2659 PubMed Central Fax: (613) 673-2462 www.pubmedcentral.com e-mail: [email protected]

Four issues of the RENEW YOUR SUBSCRIPTION Canadian Journal of Veterinary Research for 2014 NOW! (online) Quatre numéros de la RENOUVELEZ VOTRE ABONNEMENT Revue canadienne de recherche vétérinaire pour 2014 DÈS MAINTENANT! $155 $ ( applicable GST/HST; Canadian/Canadien TPS/TVH en vigueur) $170 $ US/É.-U. Foreign/Étranger $50 $ ( applicable GST/HST) Back issues/Anciens numéros each/chacun

Contact/Communiquez avec : Kelly Gray Editorial Coordinator/Coordonnatrice de la rédaction Canadian Journal of Veterinary Research Revue canadienne de recherche vétérinaire 339 rue Booth Street Ottawa, Ontario K1R 7K1 Canada Tel./Tél. : (613) 236-1162, x117 Fax/Télécopieur : (613) 236-9681 E-mail/Courriel : [email protected]

www.canadianveterinarians.net

Tables of contents Tables des matières Coming events Événements à venir Article abstracts Résumés d’articles Classified advertising Annonces classées Subscription information Information sur l’abonnement Instructions for authors Directives à l’intention des auteurs

www.veterinairesaucanada.net Specific contrast ultrasound using sterically stabilized microbubbles for early diagnosis of thromboembolic disease in a rabbit model Michal Vlašín, Robert Lukáˇc, Zuzana Kauerová, Pavel Kohout, Josef Mašek, Eliška Bartheldyová, Št ˇepán Koudelka, Zina Korvasová, Jana Plocková, Nikola Hronová, Jaroslav Turánek ...... 133

Short Communications/ Communications brèves Seroprevalence of feline immunodeficiency virus (FIV) and feline leukemia virus (FeLV) in shelter cats on the island of Newfoundland, Canada Hannah J. Munro, Lesley Berghuis, Andrew S. Lang, Laura Rogers, Hugh Whitney ...... 140

Antimicrobial susceptibility of Streptococcus suis isolated from clinically healthy swine in Brazil Taíssa Cook Siqueira Soares, Antonio Carlos Paes, Jane Megid, Paulo Eduardo Martins Ribolla, Karina dos Santos Paduan, Marcelo Gottschalk ...... 145

Evaluation of a real-time polymerase chain reaction assay of the outer membrane protein P2 for the detection of Haemophilus parasuis in clinical samples Rebeccah McDowall, Durda Slavic, Janet I. MacInnes, Hugh Y. Cai ...... 150

Effect of native Lactobacillus murinus LbP2 administration on total fecal IgA in healthy dogs Luis Delucchi, Martín Fraga, Karen Perelmuter, Claudia Della Cella, Pablo Zunino ...... 153

Ex vivo evaluation of 7 polydioxanone for closure of equine ventral midline celiotomies Stacy L. Anderson, Jose L. Bracamonte, Steve Hendrick ...... 156