TJVM Vol. 48 No. 4 December 2018

608 Charoenvisal N. et al. / Thai J Vet Med. 2018. 48(4): 603-611. Discussion was not affected. The interference between IBV and NDV was previously studied by other researchers. Genotype VII NDV have been reported in Live combined ND-IB vaccine did not have any effect South East Asia countries, such as Cambodia and on IBV antibody level but had an effect on NDV Vietnam, which are neighboring countries of Thailand antibody level. This occurred due to both IBV and ( Choi et al. , 2013; Choi et al. , 2014) . Based on the NDV primarily infecting the same target cells which geographical location of these outbreaks, Thailand is at were respiratory epithelial cells and replicate in the great risk, so effective preventive measures are highly cytoplasm. Moreover, IBV vaccine can decrease recommended. Besides a strict policy of animal trading Harderian gland activity and result in a lower immune and biosecurity, an efficacious vaccination strategy is response to NDV, which cause lower NDV antibody one solution that can protect chickens from this titer as chickens in groups 1 and 2 of our study emerging virus. This study determined the efficacy of (Montgomery et al., 1997; Cardoso et al., 2005). recombinant HVT- NDV vaccine simultaneously However, all vaccinated groups had high rate of vaccinated with live vaccine at 1 day old, followed by protection against the challenge virus. All surviving a booster vaccination with live vaccine at 10 days old, chickens at 14 DPI had significantly higher NDV titers against challenge with Asiatic, Genotype VII NDV in when compared to those of 0 DPI, which were due to broiler chickens. NDV challenge. NDV vaccines are able to protect the animal The non-vaccinated, challenged control from morbidity and mortality, but chickens are still group had 95 percent mortality when challenged at 14 susceptible to the virus and are able to shed the virus. and 21 days old, and 100 percent mortality when were However, the amount of virus shedding depends on challenged at 28 and 35 days old. That is to say that, the the host’s immunity, the virulence of the virus, the viral MDA of the non- vaccinated chickens diminished load, the type of vaccine, the route of vaccination and before the chickens were 28 days old. There are several the duration between vaccination and challenge pieces of research that have monitored MDA levels ( Miller et al. , 2013) . In this study, the chickens in the against NDV. Rahman et al. (2002) reported that MDA vaccinated groups had a significantly lower mortality HI titers of up to 16 (4 log2) failed to protect the rate than the non-vaccinated challenged control group. chickens against virulent NDV, but if the HI titers were 32 (5 log2) and above, the birds were able to resist NDV The MDA were as high as 7. 00±1. 00 to infection. Jalil et al. (2009) summarized that MDA titers 8. 40±0. 55 log2 at 1 day old and declined to very low of 128 (7 log2) or above are adequate to protect chickens level (1.45±0.69 to 1.55±0.83 log2) at 21 days old. Some from virulent NDV challenge. In our study, the birds of the vaccinated groups had a lower rate of protection in group 4 had antibody titers of 2.6 ± 0.88 (log2) at 14 against early NDV challenge at 14 days old. This might days old, which was one of the challenge days. Birds in be due to the decrease of MDA to a level, which could group 4 that were challenged on that day had 95% not protect the chickens, while the antibody induced mortality. At 21 days old, the birds in group 4 had 1.45 by vaccination was still insufficient to protect them log2 antibody titers and also had 95% mortality. In against the challenge virus. As a result, there was a other words, MDA titer around 2-3 log2 is not sufficient high risk of NDV infection during 2 - 3 week of age, to protect birds from virulent NDV strains. On the when the chickens were still susceptible to the virus, other hand, chickens in the vaccinated groups had especially when the biosecurity level was not good. antibody titers around 2.05 - 4.85 log2 and exhibited significantly milder clinical signs and a lower mortality Interestingly, on the challenge date at 14 days rate. For example, chickens in group 3, challenged at 14 old, chickens in group 3 had the lowest antibody titers days old, had antibody titer of 2.40 ± 0.50 log2, which but the lowest mortality rate. The low antibody titers was lower than that of group 4. But the mortality rate at 14 days old represented the remains of passive of group 3 was only 5%. This result could be due to the antibody or MDA, since the immune response induced protective efficacy of the specific antibody induced by by the recombinant vaccine was the antibody against the inserted F protein of NDV in the recombinant the F protein of NDV, which could not be detected by vaccine. Although, NDV has 16 genotypes all of them the HI test. Moreover, the chickens’ immune system are the same serotype. So the F gene inserted in this might not yet respond to the booster vaccination of live HVT-NDV vaccine, which was from genotype II NDV, vaccine at 10 days old. Therefore, the specific antibody was able to stimulate neutralizing antibody that against the F protein of NDV, might play an important protected the chickens from genotype VII NDV role against NDV challenge. Unfortunately, the serum challenged. However, it was unable to be detected by was not tested by ELISA specific to F protein. the HI test as previously mentioned. Moreover, local Therefore, the level of antibody against the NDV F immunity and cell-mediated immunity (CMI), induced protein was not determined. by live vaccine also were not detected by the HI test, and could be additional immunity to support the high On the challenge date at 21, 28 and 35 days protection rate of those vaccinated chickens with low old, the HI antibody titers against NDV of the HI titers. In contrast, MDA is a passive immunity, vaccinated groups were significantly higher than those where the antibodies can be directly transferred to the of the non- vaccinated control groups. Among the newborns and it is in the blood circulation only. vaccinated groups, the chickens in group 3 had the highest antibody titers. This result might be due to the NDV shedding of the 35-day-old challenged difference of vaccination program for chickens in chickens was detected by real-time PCR. There was no group 3 which were not vaccinated with live NDV significant relation between the number of virus copies vaccine at 1 day old but were vaccinated with live NDV and duration of viral shedding, either in the vaccinated vaccine at 10 days old. Therefore, the interference of IBV vaccine on the antibody response of NDV vaccine

Charoenvisal N. et al. / Thai J Vet Med. 2018. 48(4): 603-611. 609 or the non-vaccinated groups (data is not shown). Choi KS, Kye SJ, Kim JY, To TL, Nguyen DT, Lee YJ, Virus shedding was detected as early as 2 DPI in both Choi JG, Kang HM, Kim KI, Song BM and Lee HS vaccinated and non- vaccinated birds. At 4 DPI, the 2014. Molecular epidemiology of Newcastle numbers of oropharyngeal swabs that were NDV disease viruses in Vietnam. Trop Anim Health positive were as high as 80% in the non- vaccinated Prod. 46: 271-277. chickens, compared to only 10- 20% in the vaccinated chickens. While the results of cloacal swabs at 4 DPI, Diel DG, Susta L, Garcia SC, Killian ML, Brown CC, were 60% NDV was positive in the non-vaccinated Miller PJ and Afonso CL 2012a. Complete groups, and 30-40% positive in the vaccinated groups. genome and clinicopathological characterization The number of chickens that were NDV positive in the of a virulent Newcastle disease virus isolate from vaccinated chickens declined at 7 DPI, while all of the south America. J Clin Microbiol. 50: 378-387. non- vaccinated chickens died before 7 DPI. Chickens in groups 1 and 2 continued to shed the virus until 14 Diel DG, Silva LH, Liu H, Wang Z, Miller PJ and DPI, while chickens in group 3 shed the virus only until Afonso CL 2012b. Genetic diversity of avian 10 DPI. Comparing groups 1-3, the least virus shedding paramyxovirus type 1: proposal for a unified group was group 3 which was 13 percent (13 positive nomenclature and classification system of samples from 100 samples tested). There were no Newcastle disease virus genotypes. Infect Genet differences between NDV positive results detected Evol. 12: 1770-1779. from oropharyngeal and cloacal swabs. Dortmans J, Koch G, Rottier PJ and Peeters BP 2011. Likewise, at 35 days old, before challenge, the Virulence of Newcastle disease virus: what is chickens in group 3 had the highest mean antibody known so far. Vet Res. 42: 122. titer levels, followed by group 2 and group 1. As mentioned earlier, vaccination could protect the Ganar K, Das M, Sinha S and Kumar S 2014. Newcastle chickens from morbidity and mortality, but did not disease virus: Current status and our protect them from infection and virus shedding (Miller understanding. Virus Res. 184: 71-81. et al. , 2013) . However, vaccination programs used in this study were able to decrease virus shedding. Hence, Hu S, Ma H, Wu Y, Liu X, Wang X, Liu Y and Liu X excellent biosecurity will further assist reducing the 2009. A vaccine candidate of attenuated genotype risks of having pathogens in farms and will aid VII Newcastle disease virus generated by reverse controlling unwanted infections. genetics. Vaccine. 27: 904-910. In conclusion, chickens which received either Jalil MA, Samad MA and Islam MT 2009. Evaluation of live C2, NDV or MA5, IBV + Clone 30, NDV vaccines maternally derived antibodies against Newcastle simultaneously vaccinated with recombinant HVT- disease virus and its effect on vaccination in NDV vaccine at 1 day old and further received a broiler chicks. Bangl J Vet Med. 7: 296-302. booster vaccination with live combined MA5, IBV + Clone 30, NDV vaccine at 10 days old, and chickens Jeon WJ, Lee EK, Lee YJ, Jeong OM, Kim YJ, Kwon JH which received recombinant HVT- NDV at 1 day old and Choi KS 2008. Protective efficacy of and were simultaneously vaccinated with live MA5, commercial inactivated Newcastle disease virus IBV and live 4-91, IBV vaccines and received a booster vaccines in chickens against a recent Korean vaccination with live Clone 30, NDV vaccine at 10 days epizootic strain. J Vet Sci. 9: 295-300. old, all had a high protection rate against high challenge level of Asiatic, Genotype VII NDV at 14, 21, Kapczynski DR and King DJ 2005. Protection of 28 and 35 days old. chickens against overt clinical disease and determination of viral shedding following Acknowledgements vaccination with commercially available Newcastle disease virus vaccines upon challenge This work was supported by grants from with highly virulent virus from the California MSD AH, Boxmeer, the Netherlands and was 2002 exotic Newcastle disease outbreak. Vaccine. supported, in part, by the Avian Health Research Unit, 23: 3424-3433. the Ratchadaphiseksomphot Endowment Fund, Chulalongkorn University. Kapczynski DR, Afonso CL and Miller PJ 2013. Immune responses of poultry to Newcastle References disease virus. Dev Comp Immunol. 41: 447-453. Cardoso WM, Aguiar Filho JLC, Romao JM, Oliveira Miller PJ, Decanini EL and Afonso CL 2010. Newcastle WF, Salles RPR, Teixeira RSC and Sobral MHR disease: evolution of genotypes and the related 2005. Effect of associated vaccines on the diagnostic challenges. Infect Genet Evol. 10: 26- interference between Newcastle disease virus 35. and infectious bronchitis virus in broilers. Rev Bras Cienc Avic. 7: 181-184. Miller PJ, Afonso CL, Attrache JE, Dorsey KM, Courtney SC, Guo Z and Kapczynski DR 2013. Choi KS, Kye SJ, Kim JY, Damasco SS, Lee YJ, Choi JG, Effects of Newcastle disease virus antibodies on Kang HM, Kim KI, Song BM and Lee HS 2013. the shedding and transmission of challenge Molecular epidemiological investigation of viruses. Dev Comp Immunol. 41: 505-513. velogenic Newcastle disease viruses from village chickens in Cambodia. Virus Genes. 47: 244-249. Montgomery RD, Maslin WR and Boyle CR 1997. Effects of Newcastle disease vaccines and Newcastle disease/infectious bronchitis combination vaccines on the head-associated lymphoid tissues of the chicken. Avian Dis. 41: 399-406. OIE (World Organization of Animal Health) 2009. Newcastle disease. OIE Terrestrial Manual. Manual of diagnostic test and vaccines for terrestrial animals 2010: 576-589.

61080 Charoenvisal N. et al. / Thai J Vet Med. 2018. 48(4): 603-611. Rahman MM, Bari ASM, Giasuddin M, Islam MR, Alam J and Sil GC 2002. Evaluation of maternal and humoral immunity against Newcastle disease virus in chicken. Int J Poult Sci. 1: 161-163. Seal BS, King DJ and Sellers HS 2000. The avian response to Newcastle disease virus. Dev Comp Immunol. 24: 257-268. Suarez DL 2013. Newcastle disease, other avian paramyxoviruses, and avian metapneumovirus infections. In: Diseases of Poultry. 13th ed. D.E. Swayne (ed.). New Jersey: John Wiley & Sons, Inc. 89-107. Susta L, Miller PJ, Afonso CL and Brown CC 2011. Clinicopathological characterization in poultry of three strains of Newcastle disease virus isolated from recent outbreaks. Vet Pathol. 48: 349-360. Takada A and Hida H 1996. Protective immune response of chickens against Newcastle disease, induced by the intranasal vaccination with inactivated virus. Vet Microbiol. 50: 17-25. Umali DV, Ito H, Suzuki T, Shirota K, Katoh H and Ito T 2013. Molecular epidemiology of Newcastle disease virus isolates from vaccinated commercial poultry farms in non-epidemic areas of Japan. Virol J. 10: 330.

Charoenvisal N. et al. / Thai J Vet Med. 2018. 48(4): 603-611. 661019 บทคัดยอ่ ประสิทธภิ าพของวคั ซีนรีคอมบิแนนทเ์ ฮชวที ีเอ็นดวี ีในโปรแกรมวคั ซีนท่แี ตกตา่ งกัน ตอ่ การป้องกนั เชื้อไวรสั นิวคาสเซลิ จโี นไทป์ 7 ในไกเ่ น้ือ ณทยา เจรญิ วศิ าล1 บปุ ผา สพุ รรณโมก2 ริค คูปแมน3 จโิ รจ ศศิปรียจนั ทร์1* การศกึ ษาคร้งั น้ีเป็นการทดสอบประสิทธิภาพของวัคซีนรีคอมบแิ นนทเ์ ฮชวที เี อ็นดวี ี โดยใหพ้ ร้อมกับวัคซีนเชือ้ เป็นเมอ่ื ไกอ่ ายุ 1 วนั และให้วคั ซนี เชือ้ เปน็ ซา้ เมอื่ ไก่อายุ 10 วนั จากน้นั ให้ไวรสั นวิ คาสเซลิ จีโนไทป์ 7 ในไกเ่ นอื้ โดยแบ่งไก่เปน็ 5 กลุ่ม ไกใ่ นกลมุ่ 1, 2 และ 3 ไดร้ บั วัคซนี รคี อมบแิ นนทเ์ ฮชวที ีเอ็นดีวี และวคั ซีนเชื้อเป็นต่างชนิดกนั เมื่ออายุ 1 วนั และได้รับวัคซนี เชอื้ เปน็ ซ้าเมอ่ื อายุ 10 วัน ไกใ่ นกลุม่ 4 และ 5 เปน็ กล่มุ ควบคมุ บวกและกล่มุ ควบคมุ ลบตามลา้ ดับ ไกใ่ นกลุ่ม 1, 2, 3 และ 4 ไดร้ ับไวรสั นวิ คาสเซิล เมือ่ อายุ 14, 21, 28 และ 35 วัน จา้ นวน 20 ตวั /กล่มุ /ครง้ั สังเกตอาการป่วยและอัตราการตายเปน็ เวลา 14 วนั ภายหลงั ไกร่ บั ไวรสั แตล่ ะครั้ง และเก็บเลือดเพือ่ ตรวจแอนตบิ อดีต่อ ไวรัสนวิ คาสเซิลเมอ่ื ไกอ่ ายุ 1 วัน 10 วัน วันท่ีไกไ่ ด้รบั ไวรัส และ 14 วันภายหลงั ไก่ไดร้ บั ไวรสั นอกจากน้ยี ังได้เกบ็ ตวั อยา่ งจากไกท่ ่ีไดร้ บั ไวรสั เม่ืออายุ 35 วัน เพือ่ ตรวจหาสารพันธุกรรมของไวรสั ผลการทดลองพบว่า ไกท่ ่ไี ดร้ บั วคั ซนี และไดร้ ับไวรัสทอ่ี ายุ 14 วัน มอี ัตราการรอดชวี ิต 70-95% ส่วนกลุม่ ท่ไี ดร้ บั วคั ซีนและไวรัสทีอ่ ายุ 21, 28 หรือ 35 วัน มีอตั ราการรอดชีวติ เพิม่ ข้นึ เป็น 90-100% นอกจากน้ี ยังพบว่า กลุ่มท่ี ได้รับวคั ซีน มจี า้ นวนไกท่ ป่ี ลอ่ ยไวรสั ออกจากรา่ งกายนอ้ ยกว่า กล่มุ ที่ไม่ไดร้ บั วัคซนี ผลการทดลองนส้ี รุปไดว้ ่า โปรแกรมวคั ซีนท่ปี ระกอบด้วย วคั ซนี รีคอมบแิ นนทเ์ ฮชวีทเี อ็นดีวี ท้ัง 3 โปรแกรม สามารถลดอตั ราการสญู เสียจากไวรสั นวิ คาสเซลิ จีโนไทป์ 7 ได้ คาสาคัญ: ไก่ วัคซนี รีคอมบแิ นนท์เฮชวที ีเอน็ ดีวี ประสทิ ธิภาพ ไวรสั นิวคาสเซลิ 1หน่วยปฏิบัติการวจิ ัยสขุ ภาพสตั ว์ปกี ภาควชิ าอายุรศาสตร์ คณะสตั วแพทยศาสตร์ จฬุ าลงกรณ์มหาวิทยาลัย ปทุมวนั กรุงเทพฯ 10330 2บรษิ ทั อนิ เตอรเ์ ว็ท (ประเทศไทย) จากดั อาคารรจั นาการ ถนนสาทรใต้ แขวงยานนาวา เขตสาทร กรุงเทพฯ 10120 3เอม็ เอสดี แอนิมัล เฮลท์ บอกซ์เมยี ร์ ประเทศเนเธอรแ์ ลนด์ *ผรู้ บั ผิดชอบบทความ E-mail: [email protected]



Original Article A preliminary study on diversity of midgut microbiota in Aedes aegypti (Linnaeus) and Culex quinquefasciatus (Say) collected from Bangkok, Thailand Sonthaya Tiawsirisup1* Jirinnut Sirijutalak1 Mananya Sondang1 Manunya Supol1 Nut Ansusinha1 Ranida Tuanudom1,3 Channarong Rodkhum2 Abstract Aedes aegypti and Culex quinquefasciatus are important mosquito vectors for many infectious diseases. A number of factors affect the vector competence of these mosquitoes for a specific pathogen. The bacteria harbored in the midgut are known to influence mosquito physiology and can alter the response to various pathogens. Bacteria from the midgut of Ae. aegypti and Cx. quinquefasciatus were cultured and identified using bacteriological and molecular techniques in this study in which two groups of mosquitoes were examined. The first group was laboratory- reared, and the second group were field- collected mosquitoes from Bangkok. Twelve bacterial genera ( i. e. , Acinetobacter, Agrobacterium, Bacillus, Cellulomonas, Chryseomicrobium, Dietzia, Enterobacter, Klebsiella, Microbacterium, Pantoea, Pseudomonas, and Staphylococcus) were identified from laboratory- reared Ae. aegypti and eight bacterial genera ( i. e. , Bacillus, Cellulomonas, Microbacterium, Micrococcus, Moraxella, Neisseria, Staphylococcus, and Streptococcus) were determined from field- collected Ae. aegypti. Five bacterial genera ( i. e. , Microbacterium, Micrococcus, Paenibacillus, Pseudomonas, and Staphylococcus) were identified from laboratory- reared Cx. quinquefasciatus and 13 bacterial genera ( i. e. , Acinetobacter, Actinomyces, Bacillus, Chryseobacterium, Kocuria, Microbacterium, Micrococcus, Novosphingobium, Pantoea, Providencia, Pseudomonas, Rhodococcus, and Staphylococcus) were examined from field- collected Cx. quinquefasciatus. The variation of these midgut microbiota may influence mosquito vector competence for a specific pathogen. However, further studies need to be performed to indicate this relationship. Keywords: Bacteria, Midgut, Aedes aegypti, Culex quinquefasciatus, Thailand 1Animal Vector-Borne Disease Research Group, Parasitology Unit, Department of Veterinary Pathology, Faculty of Veterinary Science, Chulalongkorn University, Bangkok, Thailand 2Department of Veterinary Microbiology, Faculty of Veterinary Science, Chulalongkorn University, Bangkok, Thailand 3Interdisciplinary Program of Biomedical Science, Faculty of Graduate School, Chulalongkorn University, Bangkok, Thailand *Correspondence: [email protected], [email protected] Thai J Vet Med. 2018. 48(4): 613-622.

614 Tiawsirisup S. et al. / Thai J Vet Med. 2018. 48(4): 613-622. Introduction their successful development into the adult stage. However, a reduction in bacteria diversity can be Aedes aegypti (Linnaeus) and Culex found during the metamorphosis (Kim et al., 2015). quinquefasciatus (Say) are important mosquitoes that The involvement of midgut microbiota in can be found worldwide and throughout Thailand. various important functions in relation to host and They are nuisance insects and important vectors for pathogen interaction has been reported. Studies on many infectious diseases. They feed on various animal midgut microbiota diversity and the ability to hosts and humans, and sometimes bite more than one modulate host-pathogen interaction have become a host or one person within each gonotrophic cycle focus of research (Boissiere et al., 2012; Ramirez et al., (Harrington et al., 2014). Ae. aegypti exclusively feed on 2012; Apte-Deshpande et al., 2014). Some studies have suggested a potential role of microbiota in the biology humans in a single host species and most of multiple- and vector competence of mosquitoes (Dennison et al., host bloodmeals include at least one human host. 2014). The midgut microbiota can modulate the Humans, dogs and swine are preferred hosts but mosquito’s immune response and affect vector bovines and chickens are avoided as hosts for Ae. competence and can also manipulate mosquito aegypti in Thailand (Ponlawat and Harrington, 2005). competence by impairing pathogen infection through Cx. quinquefasciatus is most frequently found in and resource competition or antipathogen molecule secretion (Dennison et al., 2015). Ramirez et al. (2012) around human habitations and is most prevalent posited that field isolated bacteria from the mosquito during the rainy season. It prefers to feed on human midgut exert a harmful effect on dengue virus blood than that of other animals (Azmi et al., 2015). infection. The effect is demonstrated through the action of the mosquito immune system, which is activated by Ae. aegypti play an important role as vectors microbiota. On the other hand, dengue virus infection induces immune responses in the mosquito midgut for many filarial nematodes and viruses particularly tissue that act against the natural mosquito midgut the Zika and dengue virus (Watson and Kay, 1999; microbiota. These observations have encouraged the Tiawsirisup and Nithiuthai, 2006; Diallo et al., 2008; recent development of new mosquito control methods Ariani et al., 2015; da Moura et al., 2015; Monaghan et based on the use of symbiotically-modified mosquitoes al., 2016; Richard et al., 2016). Cx. quinquefasciatus are to interfere with pathogen transmission or reduce host also important vectors for filarial nematodes, protozoa, reproduction and life span (Minard et al., 2013). and many viruses such as canine heartworm caused by Dirofilaria immitis, avain malaria caused by Plasmodium Studying the diversity of the midgut gallinaceum, West Nile virus and Rift Valley fever virus microbiota will address the basic knowledge for the advance study of the relationship between midgut (Vargas and Beltran, 1941; Ahid et al., 2000; microbiota and specific pathogen infection and Tiawsirisup and Nithiuthai, 2006; Turell et al., 2007; transmission in the mosquito vectors. Understanding Sudeep et al., 2015; Yurayart et al., 2017). the role of microbiota in modulating infections with pathogens is important. However, information on Vertebrate hosts, pathogens and mosquito midgut microbiota of mosquitoes from Thailand is vector factors affect the vector competence of limited. This study was conducted to access the mosquitoes for a specific pathogen. Pathogen infection diversity of midgut microbiota of Ae. aegypti and Cx. in mosquitoes can occur only in the mosquito midgut quinquefasciatus from Thailand. This is basic and since the structure and function of the midgut are important information for the advanced research on different from the foregut and hindgut (Houk et al., mosquito vector competence and mosquito control in 1981; Mercado-Curiel et al., 2008). The mosquito gut is Thailand. the first point of contact between ingested pathogens and the mosquitos epithelial surface. The midgut is an Materials and Methods important location for host-pathogen interaction and pathogen survival or elimination is thought to be an Mosquitoes: Laboratory-reared and field-collected Ae. outcome of this relation. The midgut of the mosquito aegypti and Cx. quinquefasciatus were examined in this vector contains not only pathogens but also a diverse study. The laboratory mosquitoes were reared and microbiota (Dennison et al., 2014). Midgut microbiota maintained with 10% sucrose at the Parasitology Unit, are bacteria that have co-evolved and developed Department of Veterinary Pathology, Faculty of symbiotic relationships with mosquitoes or the Veterinary Science, Chulalongkorn University. They bacteria that are acquired from the mosquito’s were originally collected from Bangkok, Thailand and breeding water or nectar sources and they have maintained for more than ten generations. Field adapted to persist within the mosquitoes. These mosquitoes were collected from Suanluang and Laksi bacteria influence the mosquitoes’ physiology, the districts in Bangkok, Thailand in 2013 using a BG- susceptibility of the mosquitoes to specific pathogens, Sentinel mosquito trap (Biogents®, Germany). The the response to various pathogens, and the ability to species, source and number of tested mosquitoes is transmit the pathogen. In the same vein, internal demonstrated in Table 1. The mosquitoes were then factors of the mosquito might modulate the composition of its midgut bacterial population. Midgut kept at 4C for 20 min and identified under a light structure, pH, digestive enzymes and ingested food are microscope. Only female mosquitoes were examined factors shown to significantly influence the diversity of in this study. The sampling mosquitoes were washed the microbial community of the mosquito (Oliveira et in 70% ethanol for 5 min and rinsed with phosphate al., 2011; Ludvigsen et al., 2015). Midgut microbiota are buffer saline (PBS). The midgut was then dissected also diverse depending on species, sex, developmental stage, ecological factors, and geographic location. Mosquitoes are exposed to a variety of microbes during their lifecycle, some of which are needed for

Tiawsirisup S. et al. / Thai J Vet Med. 2018. 48(4): 613-622. 615 from each mosquito under a light microscope, kept in included among the experimental animals that needed 300 µl of 60% glycerol in PBS, ground using a sterile approval from the Chulalongkorn University Animal Care and Use Committee. plastic pestle, and kept at -80C until tested. This study was conducted in 2013, then mosquitoes were not Table 1 The species, source, and number of tested mosquitoes. Mosquito species Source of mosquitoes Number of tested Number of mosquitoes Aedes aegypti mosquitoes with midgut microbiota Laboratory Culex quinquefasciatus Suanluang district 16 13 5 4 Laksi district 7 6 Laboratory Suanluang district 11 10 Laksi district 10 7 11 11 Bacterial isolation and identification: Bacteria from -80C until being tested using polymerase chain the midguts of Ae. aegypti and Cx. quinquefasciatus were reaction and sequencing. cultured and identified using bacteriological and molecular techniques. Polymerase chain reaction technique: The extracted bacterial DNA was examined using polymerase chain Bacteriological technique: The ground midgut from reaction (PCR) technique. Two universal primers for each mosquito was separated into three parts. For the 16S ribosomal RNA (16S rRNA) gene of bacteria were first part, 50 µl of the ground midgut was examined for used in this study. The first pair of primers were 16S the total colonies using pour plate technique. The Forward 5’-AGT TTG ATC CTG GCT CAG-3’ and 16S sample was mixed with plate count agar (PCA) and Reverse 5’-GCT ACC TTG TTA CGA CTT C-3’ poured into a plastic petri dish. The petri dish was kept (Dinparast Djadid et al., 2011) and the second pair of at 37C for 24-48 hr and assessed for bacterial colonies. primer were 63F 5’-CAG GCC TAA CAC ATG CAA If the total colonies of bacteria were higher than 250 GTC-3’ and 1387R 5’-GGG CGG WGT GTA CAA colonies, the ground midgut was diluted and the total GGC-3’ (Marchesi et al., 1998). PCRs were performed colonies were examined again using pour plate. For the in 25 μl-reaction. The PCR consisted of 1.5 l of DNA second and third part, 50 µl of the ground midgut was template, 0.2 l of Taq DNA polymerase (Platinum Taq examined for bacterial colonies by spread plate DNA polymerase high fidelity, Invitrogen, USA), 2.5 l technique using trypticase soy agar (TSA) with 5% of 10X PCR buffer, 0.5 l of 10mM dNTPs, 1 l of 50 sheep blood and MacConkey agar (MAC) with 5% mM MgSO4, 10 µM of forward primer, 10 µM of reverse sheep blood being used in this technique, respectively. primer and 17.3 l of distilled water. DNA was The midgut was streaked over an agar surface by the amplified using thermocycler (Perkin Elmer Cetus four-way cross streak method and the streaked plate 9600, Perkin Elmer, Waltham, MA). was kept at 37C for 24-48 hr and assessed for bacterial colonies. The bacterial colonies were indicated as Reaction for the first primer consisted of the colony forming units (CFU) per mosquito. initial denaturation at 94C for 2 min, the amplification was carried out for 35 cycles with the following Each bacterial colony from the spread plate temperature cycling parameters: 94C for 30 s of technique was subcultured over TSA with 5% sheep denaturation, 55C for 30 s of annealing, and 68C for blood by four-way cross streak method and the 1 min 30 s of extension. The final amplification cycle streaked plate was kept at 37C for 24-48 hr. Pure included an addition of 10 min extension at 72C. bacterial colonies were examined using gram staining Reaction for the second primer consisted of the initial and molecular techniques. Before examination by denaturation at 94C for 2 min, the amplification was molecular technique, the pure bacterial colony was carried out for 30 cycles with the following cultured in Luria-Bertani (LB) broth and stock of the temperature cycling parameters: 95C for 1 min of pure colony was kept in stock media at 37C for 24 hr denaturation, 55C for 1 min of annealing, and 72C for and transferred to room temperature. Duplicate 1 min 30 s of extension. The final amplification cycle samples of each mosquito midgut were analyzed. included an addition of 5 min extension at 72C. The PCR product was mixed with loading buffer Molecular technique: Due to the various sizes and (BlueJuice™ Gel Loading Buffer, Invitrogen, USA) and shapes of the colony of each bacterial genus, the analyzed in 1.5% agarose gel (UltraPure™, Invitrogen, molecular technique using DNA sequencing was Carlsbad, CA) with an expected 1.5 and 1.3 kilobase mainly used to indicate the bacterial genus. Bacterial pair band, respectively. DNA was extracted using the boiling method. The pure bacterial colony was cultured in two ml of LB Bacterial DNA sequencing: After DNA amplification, broth at 37C for 24-48 hr. It was then centrifuged at the PCR product in agarose gel was purified using 14,000 rpm for 15 min and the bacterial pellet was Gel/PCR DNA Fragments Extraction Kit (Geneaid, washed one time in distilled water. A total of 40 µl of Taiwan) according to the manufacturer’s distilled water was added into the bacterial pellet and recommendation. The purified DNA was sequenced it was kept at 100C for 10 min, cooled down in the ice (First BASE Laboratories, Singapore), analysed using basket, and centrifuged at 14,000 rpm for 10 min. The Molecular Evolution Genetics Analysis (MEGA) 5.1, supernatant which was extracted DNA was kept at and blasted with the data in GenBank.

616 Tiawsirisup S. et al. / Thai J Vet Med. 2018. 48(4): 613-622. Results and 12 bacterial genera (i.e., Acinetobacter, Agrobacterium, Bacillus, Cellulomonas, Chryseomicrobium, This study was conducted to examine the Dietzia, Enterobacter, Klebsiella, Microbacterium, Pantoea, genera and diversity of bacteria in the midguts of Ae. Pseudomonas, and Staphylococcus) were identified from aegypti and Cx. quinquefasciatus mosquitoes. Laboratory-reared and field-collected mosquitoes this mosquito group (13/16). The number of bacterial were examined in this study. The field mosquitoes colonies ranged from 1-142 CFU/mosquito. Three were collected from Bangkok, Thailand using BG- mosquitoes were free from the culturable bacteria in Sentinel mosquito traps. Bacteria in the mosquito the midguts (3/16). The most common bacteria found midguts were cultured and identified by using in this mosquito group were Microbacterium (4/16). bacteriological and molecular techniques. A total of 22 genera were identified, belonging to 4 phyla: Proteobacteria was the dominant bacterial phylum Actionobacteria, Bacteroidetes, Firmicutes, and followed by Actinobacteria and Firmicutes, Proteobacteria. Proteobacteria was the dominant respectively. bacterial phylum followed by Actinobacteria, Firmicutes, and Bacteroidetes, respectively. Found in Twelve field-collected Ae. aegypti from this study by using the bacteriological technique, some bacterial colonies could not be identified using Bangkok were examined and eight bacterial genera molecular technique. This was due to the limitation of (i.e., Bacillus, Cellulomonas, Microbacterium, Micrococcus, the boiling method that was used for bacterial DNA Moraxella, Neisseria, Staphylococcus, and Streptococcus) extraction. were identified from this mosquito group (10/12). The Bacterial isolates from the midgut of Ae. Aegypti: number of bacterial colonies ranged from 1-2,900 Sixteen laboratory-reared Ae. aegypti were examined CFU/mosquito. Two mosquitoes were free from the culturable bacteria in the midgut (2/12). The most common bacteria found in this mosquito group were Staphylococcus (6/12). Actinobacteria and Firmicutes were the dominant bacterial phyla followed by Proteobacteria (Table 2 and 4). Table 2 Number of bacterial colonies found in laboratory-reared and field-collected Aedes aegypti from Bangkok, Thailand. ID Source No. of bacterial colonies per mosquito (CFU) Closest related bacterial genera 1 Laboratory Pour plate MAC plate TSA plate 2 Laboratory Microbacterium 3 Laboratory 70 8 Enterobacter Pantoea 4 Laboratory 19 0 1 Klebsiella 5 Laboratory Acinetobacter 6 Laboratory 80 2 Agrobacterium 7 Laboratory - 8 Laboratory 03 Bacillus 9 Laboratory 02 Microbacterium 10 Laboratory Dietzia cinnamea 11 Laboratory 02 Microbacterium 12 Laboratory 13 Laboratory 40 1 Pseudomonas Chryseomicrobium 14 Laboratory 20 4 15 Laboratory - 16 Laboratory 30 4 17 Field (Suanluang) Bacillus 18 Field (Suanluang) 10 1 Cellulomonas 19 Field (Suanluang) 20 0 14 - 20 Field (Suanluang) 21 Field (Suanluang) 05 Staphylococcus 22 Field (Laksi) Pseudomonas 01 Microbacterium 23 Field (Laksi) Staphylococcus 10 5 Staphylococcus 24 Field (Laksi) 25 Field (Laksi) 00 3 - 26 Field (Laksi) 11 0 12 Staphylococcus 27 Field (Laksi) Staphylococcus 28 Field (Laksi) 30 30 Micrococcus 00 19 Bacillus 01 Micrococcus Micrococcus 10 142 Moraxella 50 18 Microbacterium Staphylococcus 40 13 Cellulomonas 00 1 Staphylococcus Staphylococcus 10 1 Neisseria 50 1 Streptococcus 02 - 40 33 Staphylococcus 30 1 00 5 0 11 10 1 02 10 1 40 2 470 0 2,500 0 700 0 2,900 10 1 10 41

Tiawsirisup S. et al. / Thai J Vet Med. 2018. 48(4): 613-622. 617 Bacterial isolates from the midgut of Cx. Twenty-one field-collected Cx. quinquefasciatus: Eleven laboratory-reared Cx. quinquefasciatus were examined and five bacterial quinquefasciatus from Bangkok were examined and 13 genera (i.e., Microbacterium, Micrococcus, Paenibacillus, Pseudomonas, and Staphylococcus) were identified from bacterial genera (i.e., Acinetobacter, Actinomyces, this mosquito group (10/11). The number of bacterial Bacillus, Chryseobacterium, Kocuria, Microbacterium, colonies ranged from 1-140 CFU/mosquito. One mosquito was free from the culturable bacteria in the Micrococcus, Novosphingobium, Pantoea, Providencia, midgut (1/11). The most common bacteria found in this mosquito group were Staphylococcus (6/11). Pseudomonas, Rhodococcus, and Staphylococcus) were Actinobacteria was the dominant bacterial phylum identified from this mosquito group (18/21). The followed by Bacteroidetes, Firmicutes, and Proteobacteria. number of bacterial colonies ranged from 1-18,800 CFU/mosquito. Three mosquitoes were free from the culturable bacteria in the midgut (3/21). The most common bacteria found in this mosquito group were Micrococcus (5/21). Proteobacteria was the dominant bacterial phylum followed by Actinobacteria and Firmicutes, respectively (Table 3 and 4). Table 3 Number of bacterial colonies found in laboratory-reared and field-collected Culex quinquefasciatus from Bangkok, Thailand. ID Source No. of bacterial colonies per mosquito (CFU) Closest related 1 Laboratory bacterial genera 2 Laboratory Pour plate MAC plate TSA plate 3 Laboratory - 10 70 4 Laboratory Micrococcus 40 35 Staphylococcus 5 Laboratory 6 Laboratory 40 11 Micrococcus Microbacterium 7 Laboratory 01 Microbacterium 8 Laboratory Staphylococcus 0 40 Staphylococcus 9 Laboratory 20 0 9 Paenibacillus 10 Laboratory Microbacterium 11 Laboratory 01 Microbacterium 12 Field (Suanluang) 10 1 Paenibacillus 13 Field (Suanluang) Microbacterium 14 Field (Suanluang) 96 0 2 Staphylococcus Pseudomonas 15 Field (Suanluang) 0 140 Staphylococcus 16 Field (Suanluang) Staphylococcus 17 Field (Suanluang) 00 1 18 Field (Suanluang) Providencia 19 Field (Suanluang) 14 0 3 Pantoea 20 Field (Suanluang) 0 62 Micrococcus - 21 Field (Suanluang) 30 5 22 Field (Laksi) - 23 Field (Laksi) 10 Chryseobacterium 24 Field (Laksi) 26 0 2 - 25 Field (Laksi) 26 Field (Laksi) 30 3 Pantoea 27 Field (Laksi) Micrococcus 28 Field (Laksi) 608 1,400 18,800 Staphylococcus Microbacterium 29 Field (Laksi) 240 0 Micrococcus 30 Field (Laksi) Microbacterium 28 0 7 31 Field (Laksi) - 52 0 56 32 Field (Laksi) Pseudomonas 18 0 Acinetobacter Staphylococcus 00 1 Pseudomonas Actinomyces 288 86 1,420 Staphylococcus Novosphingobium 224,000 394 0 Pantoea 20 0 1 Staphylococcus 40 1 Kocuria Micrococcus 01 Bacillus 50 1 Kocuria Pantoea 50 1 Microbacterium Bacillus 00 1 Rhodococcus 00 1 30 2 09 01 10 1 12 0 17 50 2 91 0 82 40 1 01 04 30 1 10 9 08 60 1 01 50 1

618 Tiawsirisup S. et al. / Thai J Vet Med. 2018. 48(4): 613-622. Table 4 Comparison of the occurrence of different bacterial genera in laboratory-reared and field-collected Aedes aegypti and Culex quinquefasciatus from Bangkok, Thailand. Closest related Percentage of occurrence bacterial generaa Phylum Laboratory-reared Field-collected Laboratory-reared Field-collected Ae. aegypti Cx. quinquefasciatus Ae. aegypti Cx. quinquefasciatus 4.8 (1/21) Actinobacteria Actinomyces 00 0 0 0 Cellulomonas 6.3 (1/16) 8.3 (1/12) 0 9.5 (2/21) Dietzia 6.3 (1/16) 0 0 14.3 (3/21) 23.8 (5/21) Kocuria 00 0 4.8 (1/21) Microbacterium 25.0 (4/16) 8.3 (1/12) 45.5 (5/11) 0 9.5 (2/21) Micrococcus 0 25.0 (3/12) 18.2 (2/11) 19.0 (4/21) Rhodococcus 00 0 0 4.8 (1/21) Bacteroidetes Paenibacillus 0 0 18.2 (2/11) 0 Firmicutes Bacillus 12.5 (2/16) 8.3 (1/12) 0 4.8 (1/21) Staphylococcus 18.8 (3/16) 50.0 (6/12) 54.5 (6/11) 0 Streptococcus 0 8.3 (1/12) 0 0 Proteobacteria Acinetobacter 6.3 (1/16)b 0 0 0 Agrobacterium 6.3 (1/16) 0 0 0 Chryseobacterium 6.3 (1/16) 0 0 4.8 (1/21) Enterobacter 6.3 (1/16) 0 0 19.0 (4/21) Klebsiella 6.3 (1/16) 0 0 4.8 (1/21) Moraxella 0 8.3 (1/12) 0 9.5 (2/21) Neisseria 0 8.3 (1/12) 0 Novosphingobium 0 0 0 Pantoea 6.3 (1/16) 0 0 Providencia 00 0 Pseudomonas 12.5 (2/16) 0 9.1 (1/11) a All bacterial genera were identified on the basis of a percent identity higher than 99% b Percentage of occurrence (no. occurred/tested) Discussion laboratory-reared female Cx. quinquefasciatus and 13 The mosquito midgut is a site of complex bacterial genera were examined from field-collected interactions among mosquitoes, pathogens, and female Cx. quinquefasciatus. The most common bacteria resident microbiota. The variation of these bacteria found in laboratory-reared and field-collected Ae. may influence mosquito biology and vector aegypti were Microbacterium and Staphylococcus, competence for a specific pathogen (Chandel et al., 2013; Minard et al., 2013). It is one of the factors respectively and the most common bacteria found in responsible for the difference in disease transmission laboratory-reared and field-collected Cx. rate or vector competence within the mosquito quinquefasciatus were Staphylococcus and Micrococcus, population. Previous studies indicate the role and relationship between mosquito midgut microbiota and respectively. The difference in bacterial genera vector competence for specific pathogens (Dennison et identified from each field-collected mosquito might be al., 2014). However, the information about midgut attributable to environmental conditions. microbiota and the relationship between these bacteria and pathogens from Thailand is limited. This study Chandel et al. (2013) studied the midgut was performed to initiate basic information about microbiota of female Cx. quinquefasciatus mosquitoes midgut microbiota from Thailand’s mosquitoes. Laboratory mosquitoes were originally collected from collected from India. The 16S ribosomal DNA from Bangkok, Thailand and maintained for more than ten culturable microflora were examined and revealed the generations and field mosquitoes were also collected presence of 83 bacterial species belonging to 31 genera. from Bangkok to indicate the effect of environmental Proteobacteria was the most dominant phylum, conditions on the variation of midgut microbiota. followed by Firmicutes and Actinobacteria. Molecular analysis of the 16S ribosomal RNA gene of Staphylococcus was the largest genus represented by 11 bacteria was used for bacterial identification in this species whereas Enterobacter was the most prevalent study. Individual mosquitoes harbor extremely diverse gut bacteria in their gut. All of the laboratory- genus and recovered from most field stations. reared mosquitoes in this study were raised and fed in However, only 13 bacterial genera were identified the same manner. However, the midgut microbiota from field-collected Cx. quinquefasciatus from Thailand identified from these mosquitoes was different. The microbiota in each generation of mosquitoes was likely in this study. Chandel et al. (2015) also isolated to become more diverse during the course of the Vagococcus fluvialis from Cx. quinquefasciatus mosquito experiment in laboratory condition. midgut collected from India where these bacteria were Twelve bacterial genera were identified from known from domestic animal and human sources only. laboratory-reared female Ae. aegypti and eight bacterial This finding might confirm the hypothesis that genera were determined from field-collected female microbiota is acquired from food sources of the Ae. aegypti. Five bacterial genera were identified from mosquito (Ludvigsen et al., 2015). However, no Vagococcus was isolated from our present study from Thailand. The study by Valiente Moro et al. (2013) showed the bacterial isolates from female Ae. albopictus mosquitoes were mostly Proteobacteria followed by Firmicutes and Actinobacteria phylum. On the other hand, Actinobacteria was the most abundant phylum

Tiawsirisup S. et al. / Thai J Vet Med. 2018. 48(4): 613-622. 619 in male Ae. albopictus followed by Proteobacteria and genera identified from different mosquito species and Firmicutes. Pantoea was the most common genus in locations were caused by both external factors (e.g., environmental conditions) and internal factors. This both females and males from all sampling sites. In our microbiota diversity could indicate the differences in present study, Pantoea was isolated from laboratory- vector competence among mosquito species and reared female Ae. aegypti and field-collected female Cx. strains. Future studies of the role of culturable bacteria quinquefasciatus from Thailand. The study of midgut in the biological role in the invasiveness of Ae. aegypti microbiota of Ae. albopictus and Ae. aegypti collected and Cx. quinquefasciatus need to be performed. Isolated from India by Yadav et al. (2015) found 24 bacterial bacteria should be characterized to better understand species from 13 genera of four major phyla using 16S its genetic contents and any possible to influence on Ae. rRNA gene sequence analysis. Phylum Proteobacteria aegypti and Cx. quinquefasciatus vector competence. was dominant followed by Firmicutes, Bacteroidetes, and Actinobacteria. The bacteria belonging to the Acknowledgements phylum Proteobacteria and Firmicutes were identified from both Ae. albopictus and Ae. aegypti, while, bacteria This study was supported by the Thailand Research Fund and Chulalongkorn University (RSA belonging to phylum Bacteroidetes and Actinobacteria 5680030), Special Task Force for Activating Research, were isolated only from Ae. albopictus and Ae. aegypti, Chulalongkorn University (GSTAR 59-001-31-001), respectively. Enterobacter was the dominant bacterial and the Faculty of Veterinary Science, Chulalongkorn genus in both Ae. albopictus and Ae. aegypti. The same University. result was found in this present study where bacteria References in phylum Bacteroidetes was only found in laboratory- reared female Cx. quinquefasciatus. Ahid SM, Vasconcelos PS and Lourenco-de-Oliviera R 2000. Vector competence of Culex quinquefasciatus For the midgut microbiota of Anopheles Say from different regions of Brazil to Dirofilaria immitis. Mem Inst Oswaldo Cruz. 95(6): 769-775. mosquitoes, the majority of the identified bacteria from An. stephensi and An. maculipennis from Iran belonged Apte-Deshpande AD, Paingankar MS, Gokhale MD to proteobacteria phylum, including Pseudomonas and and Deobagkar DN 2014. Serratia odorifera Aeromonas. The other bacteria found in these mediated enhancement in susceptibility of Aedes mosquitoes were Pantoea, Acinetobacter, Brevundimonas, aegypti for chikungunya virus. Indian J Med Res. Bacillus, Sphingomonas, Lysinibacillus, and Rahnella 139(5): 762-768. (Dinparast Djadid et al., 2011). Pantoea agglomerans has anti-Plasmodium effector proteins that reduce mosquito Ariani CV, Juneja P, Smith S, Tinsley MC and Jiggins refractoriness to malaria infection and engineered P. FM 2015. Vector competence of Aedes aegypti agglomerans strains are able to inhibit Plasmodium mosquitoes for filarial nematodes is affected by falciparum development (Wang et al., 2012). Other age and nutrient limitation. Exp Gerontol. 61: 47- 53. studies showed mosquitoes with an important microbiota seem more resistant to infections and Azmi SA, Das S and Chatterjee S 2015. Seasonal certain bacteria, such as Enterobacter partially or totally prevalence and blood meal analysis of filarial vector Culex quinquefasciatus in coastal areas of and inhibit ookinete, oocyst, and sporozoite formation Digha, West Bengal, India. J Vector Borne Dis. (Cirimotich et al., 2011). Co-infection between Serratia 52(3): 252-256. marcescens and Plasmodium vivax in An. albimanus Boissiere A, Tchioffo MT, Bachar D, Abate L, Marie A, resulted in only 1% of mosquitoes being infected with Nsango SE, Shahbazkia HR, Awono-Ambene PH, oocysts (Gonzalez-Ceron et al., 2003). Acinetobacter, Levashina EA, Christen R and Morlais I 2012. Bacillus, Enterobacter, and Pseudomonas are also found Midgut microbiota of the malaria mosquito vector Anopheles gambiae and interactions with in the mosquitoes from Thailand which was indicated Plasmodium falciparum infection. PLoS Pathog. by the present observation. 8(5): e1002742. Another important bacterial endosymbiont in Chandel K, Mendki MJ, Parikh RY, Kulkarni G, Tikar mosquitoes is Wolbachia (Tiawsirisup et al., 2008). SN, Sukumaran D, Prakash S, Parashar BD, Shouche YS and Veer V 2013. Midgut microbial These bacteria block the transmission of arbovirus by community of Culex quinquefasciatus mosquito Aedes mosquitoes and this is currently being evaluated populations from India. PLoS One 8(11): e80453. for the control of the disease outbreak (Mousson et al., Chandel K, Parikh RY, Mendki MJ, Shouche YS and 2010; van den Hurk et al., 2012; Hussain et al., 2013; Veer V 2015. Isolation and characterization of Raquin et al., 2015). Wolbachia induces cytoplasmic Vagococcus sp from midgut of Culex quinquefasciatus (Say) mosquito. J Vector Borne incompatibility that results in the developmental Dis. 52(1): 52-57. failure of offspring (Segoli et al., 2014). However, only agar-culturable bacteria were examined in this study. Cirimotich CM, Dong Y, Clayton AM, Sandiford SL, Wolbachia infection in Ae. aegypti and Cx. Souza-Neto JA, Mulenga M and Dimopoulos G quinquefasciatus cannot be indicated by the 2011. Natural microbe-mediated refractoriness to Plasmodium infection in Anopheles gambiae. Science methodology used in this study. In addition, the 332(6031): 855-858. presence of antibiotics in the blood of infected people presents a new risk of disease transmission increasing. Antibiotics in malaria-infected people enhances the susceptibility of An. gambiae to malaria infection by disturbing their midgut microbiota. Antibiotics exposure also increases mosquito survival and fecundity, which are known to increase vector competency (Gendrin et al., 2015). Diverse bacterial

620 Tiawsirisup S. et al. / Thai J Vet Med. 2018. 48(4): 613-622. da Moura AJ, de Melo Santos MA, Oliveira CM, 16S rRNA. Appl Environ Microbiol. 64(2): 795- Guedes DR, de Carvalho-Leandro D, da Cruz 799. Brito ML, Rocha HD, Gomez LF and Ayres CF Mercado-Curiel RF, Black WCT and Munoz Mde L 2015. Vector competence of the Aedes aegypti 2008. A dengue receptor as possible genetic marker of vector competence in Aedes aegypti. population from Santiago Island, Cape Verde, to different serotypes of dengue virus. Parasit BMC Microbiol. 8: 118. Vectors. 8: 114. Minard G, Mavingui P, Moro CV 2013. Diversity and Dennison NJ, BenMarzouk-Hidalgo OJ and Dimopoulos G 2015. MicroRNA-regulation of function of bacterial microbiota in the mosquito Anopheles gambiae immunity to Plasmodium holobiont. Parasit Vectors. 6: 146. falciparum infection and midgut microbiota. Dev Monaghan AJ, Morin CW, Steinhoff DF, Wilhelmi O, Hayden M, Quattrochi DA, Reiskind M, Lloyd Comp Immunol. 49(1): 170-178. AL, Smith K, Schmidt CA, Scalf PE and Ernst K Dennison NJ, Jupatanakul N and Dimopoulos G 2014. 2016. On the seasonal occurrence and abundance of the Zika virus vector mosquito Aedes aegypti in The mosquito microbiota influencesvector competence for human pathogens. Curr Opin the contiguous United States. PLoS Curr. 2016 Insect Sci. 3: 6-13. Mar 16; 8. Diallo M, Ba Y, Faye O, Soumare ML, Dia I and Sall AA Mousson L, Martin E, Zouache K, Madec Y, Mavingui 2008. Vector competence of Aedes aegypti P and Failloux AB 2010. Wolbachia modulates Chikungunya replication in Aedes albopictus. Mol populations from Senegal for sylvatic and Ecol. 19(9): 1953-1964. epidemic dengue 2 virus isolated in West Africa. Oliveira JH, Goncalves RL, Lara FA, Dias FA, Gandara Trans R Soc Trop Med Hyg. 102(5): 493-498. AC, Menna-Barreto RF, Edwards MC, Laurindo Dinparast Djadid N, Jazayeri H, Raz A, Favia G, Ricci I FR, Silva-Neto MA, Sorgine MH and Oliveira PL and Zakeri S 2011. Identification of the midgut 2011. Blood meal-derived heme decreases ROS microbiota of An. stephensi and An. maculipennis levels in the midgut of Aedes aegypti and allows for their application as a paratransgenic tool proliferation of intestinal microbiota. PLoS against malaria. PLoS One. 6(12): e28484. Pathog. 7(3): e1001320. Gendrin M, Rodgers FH, Yerbanga RS, Ouedraogo JB, Ponlawat A and Harrington LC 2005. Blood feeding Basanez MG, Cohuet A and Christophides GK patterns of Aedes aegypti and Aedes albopictus in 2015. Antibiotics in ingested human blood affect the mosquito microbiota and capacity to transmit Thailand. J Med Entomol. 42(5): 844-849. malaria. Nat Commun. 6: 5921. Ramirez JL, Souza-Neto J, Torres Cosme R, Rovira J, Gonzalez-Ceron L, Santillan F, Rodriguez MH, Mendez D and Hernandez-Avila JE 2003. Bacteria Ortiz A, Pascale JM and Dimopoulos G 2012. in midguts of field-collected Anopheles albimanus Reciprocal tripartite interactions between the block Plasmodium vivax sporogonic development. Aedes aegypti midgut microbiota, innate immune J Med Entomol. 40(3): 371-374. system and dengue virus influences vector Harrington LC, Fleisher A, Ruiz-Moreno D, Vermeylen competence. PLoS Negl Trop Dis. 6(3): e1561. Raquin V, Valiente Moro C, Saucereau Y, Tran FH, F, Wa CV, Poulson RL, Edman, JD, Clark JM, Potier P and Mavingui P 2015. Native Wolbachia Jones JW, Kitthawee S and Scott TW 2014. from Aedes albopictus blocks Chikungunya virus Heterogeneous feeding patterns of the dengue vector, Aedes aegypti, on individual human hosts infection in cellulo. PLoS One. 10(7): e0125066. Richard V, Paoaafaite T and Cao-Lormeau VM 2016. in rural Thailand. PLoS Negl Trop Dis. 8(8): e3048. Vector competence of Aedes aegypti and Aedes Houk EJ, Hardy JL and Chiles RE 1981. Permeability of polynesiensis populations from French polynesia the midgut basal lamina in the mosquito, Culex tarsalis Coquillett (Insecta, Diptera). Acta Trop. for Chikungunya virus. PLoS Negl Trop Dis. 10(5): e0004694. 38(2): 163-171. Segoli M, Hoffmann AA, Lloyd J, Omodei GJ and Hussain M, O'Neill SL and Asgari S 2013. Wolbachia Ritchie SA 2014. The effect of virus-blocking Wolbachia on male competitiveness of the dengue interferes with the intracellular distribution of vector mosquito, Aedes aegypti. PLoS Negl Trop Argonaute 1 in the dengue vector Aedes aegypti by Dis. 8(12): e3294. manipulating the host microRNAs. RNA Biol. Sudeep AB, Mandar P, Ghodke YK, George RP and 10(12): 1868-1875. Kim CH, Lampman RL and Muturi EJ 2015. Bacterial Gokhale MD 2015. Vector competence of two communities and midgut microbiota associated Indian populations of Culex quinquefasciatus with mosquito populations from waste tires in East-Central Illinois. J Med Entomol. 52(1): 63-75. (Diptera: Culicidae) mosquitoes to three West Ludvigsen J, Rangberg A, Avershina E, Sekelja M, Nile virus strains. J Vector Borne Dis. 52(3): 185- Kreibich C, Amda G and Rudi K 2015. Shifts in 192. the midgut/pyloric microbiota composition Tiawsirisup S and Nithiuthai S 2006. Vector within a honey bee apiary throughout a season. competence of Aedes aegypti (L.) and Culex Microbes Environ. 30(3): 235-244. quinquefasciatus (Say) for Dirofilaria immitis Marchesi JR, Sato T, Weightman AJ, Martin TA, Fry JC, Hiom SJ, Dymock D and Wade WG 1998. Design (Leidy). Southeast Asian J Trop Med Public and evaluation of useful bacterium-specific PCR Health. 37(Suppl 3): 110-114. primers that amplify genes coding for bacterial Tiawsirisup S, Sripatranusorn S, Oraveerakul K, Nuchprayoon S 2008. Distribution of mosquito (Diptera: Culicidae) species and Wolbachia (Rickettsiales: Rickettsiaceae) infections during

Tiawsirisup S. et al. / Thai J Vet Med. 2018. 48(4): 613-622. 62119 1 the bird immigration season in Pathumthani province, central Thailand. Parasitol Res. 102(4): 731-735. Turell MJ, Lee JS, Richardson JH, Sang RC, Kioko EN, Agawo MO, Pecor J and O'Guinn ML 2007. Vector competence of Kenyan Culex zombaensis and Culex quinquefasciatus mosquitoes for Rift Valley fever virus. J. Am. Mosq. Control Assoc. 23(4): 378-382. Valiente Moro C, Tran FH, Raharimalala FN, Ravelonandro P and Mavingui P 2013. Diversity of culturable bacteria including Pantoea in wild mosquito Aedes albopictus. BMC Microbiol. 13: 70. van den Hurk AF, Hall-Mendelin S, Pyke AT, Frentiu FD, McElroy K, Day A, Higgs S and O'Neill SL 2012. Impact of Wolbachia on infection with chikungunya and yellow fever viruses in the mosquito vector Aedes aegypti. PLoS Negl Trop Dis. 6: e1892. Vargas L and Beltran E 1941. Culex quinquefasciatus, a new vector of Plasmodium gallinaceum. Science. 94(2443): 389-390. Wang S, Ghosh AK, Bongio N, Stebbings KA, Lampe DJ and Jacobs-Lorena M 2012. Fighting malaria with engineered symbiotic bacteria from vector mosquitoes. Proc Natl Acad Sci USA. 109(31): 12734-12739. Watson TM and Kay BH 1999. Vector competence of Aedes notoscriptus (Diptera: Culicidae) for Barmah Forest virus and of this species and Aedes aegypti (Diptera: Culicidae) for dengue 1-4 viruses in Queensland, Australia J Med Entomol. 36(4): 508- 514. Yadav KK, Bora A, Datta S, Chandel K, Gogoi HK, Prasad GB and Veer V 2015. Molecular characterization of midgut microbiota of Aedes albopictus and Aedes aegypti from Arunachal Pradesh, India. Parasi. Vectors. 8: 641. Yurayart N, Kaewthamasorn M and Tiawsirisup S 2017. Vector competence of Aedes albopictus (Skuse) and Aedes aegypti (Linnaeus) for Plasmodium gallinaceum infection and transmission. Vet Parasitol. 241: 20-25.

6202 Tiawsirisup S. et al. / Thai J Vet Med. 2018. 48(4): 613-622. 1 บทคัดยอ่ การศกึ ษาเบือ้ งตน้ เก่ียวกบั ความหลากหลายของแบคทเี รียในทางเดินอาหารสว่ นกลาง ของยุงลายบ้านและยงุ ราคาญจากเขตกรุงเทพมหานคร ประเทศไทย สนธยา เตียวศิริทรัพย์1* จริ นิ ทร์นุช สริ ิจฑุ าลกั ษณ์1 มนัญญา สนแดง1 มนัณญา สุผล1 ณฏั ฐ์ องั ศสุ ิงห์1 รนดิ า ต่วนอุดม1,3 ชาญณรงค์ รอดคา2 ยุงลายบา้ นและยงุ ราคาญเปน็ แมลงพาหะทีน่ าเชอ้ื ท่ีสาคญั หลายชนดิ ปัจจัยที่มผี ลตอ่ ศักยภาพของยุงในการนาเชอ้ื นัน้ มหี ลายอยา่ ง แบคทีเรียในทางเดินอาหารสว่ นกลางของยุงน้ันมีผลตอ่ ลกั ษณะทางสรีรวิทยาของยงุ และมีบทบาทในการเปลย่ี นแปลงการตอบสนองต่อเชื้อ โรคของยงุ การศกึ ษานเี้ ปน็ การตรวจหาชนิดของเชอ้ื แบคทเี รยี จากทางเดนิ อาหารส่วนกลางของยุงลายบา้ นและยงุ ราคาญโดยวิธีทางแบคทีเรีย วิทยาและอณชู ีววทิ ยา ตัวอยา่ งยุงทน่ี ามาศกึ ษาแบง่ ออกเปน็ 2 กลมุ่ คอื กลุ่มท่ีหนึ่งเป็นตวั อย่างยุงทไี่ ดจ้ ากการเพาะเลีย้ งในห้องปฏิบัติการ และกลมุ่ ทีส่ องเปน็ ตวั อยา่ งยงุ ทเี่ ก็บมาจากพ้ืนท่ีในกรงุ เทพมหานคร การศกึ ษานต้ี รวจพบเช้ือแบคทเี รียจานวน 12 สกุล ไดแ้ ก่ Acinetobacter, Agrobacterium, Bacillus, Cellulomonas, Chryseomicrobium, Dietzia, Enterobacter, Klebsiella, Microbacterium, Pantoea, Pseudomonas และ Staphylococcus จากยงุ ลายบ้านทเ่ี พาะเลย้ี งในห้องปฏบิ ัติการ และตรวจพบเช้ือ แบคทเี รยี จานวน 8 สกุล ไดแ้ ก่ Bacillus, Cellulomonas, Microbacterium, Micrococcus, Moraxella, Neisseria, Staphylococcus และ Streptococcus จากยงุ ลายบ้านท่ีเก็บตวั อยา่ งจากพน้ื ทใ่ี นกรงุ เทพมหานคร สาหรับยุงราคาญทเ่ี พาะเล้ียงในหอ้ งปฏบิ ัตกิ ารน้นั ตรวจพบ เช้อื แบคทเี รยี จานวน 5 สกุล ไดแ้ ก่ Microbacterium, Micrococcus, Paenibacillus, Pseudomonas และ Staphylococcus และยงุ ราคาญท่เี ก็บตัวอยา่ งจากพน้ื ทีใ่ นกรงุ เทพมหานครนนั้ ตรวจพบเช้ือแบคทเี รียจานวน 13 สกุล ได้แก่ Acinetobacter, Actinomyces, Bacillus, Chryseobacterium, Kocuria, Microbacterium, Micrococcus, Novosphingobium, Pantoea, Providencia, Pseudomonas, Rhodococcus และ Staphylococcus ความหลากหลายของเชื้อแบคทเี รยี ทต่ี รวจพบในทางเดนิ อาหารส่วนกลางของยงุ อาจมผี ลต่อศักยภาพของยงุ ในการนาเชื้อชนดิ ต่างๆ อย่างไรก็ตามจาเป็นตอ้ งมีการศึกษาเพม่ิ เติมเพื่อบ่งชถี้ ึงความสัมพนั ธ์น้ี คาสาคญั : แบคทีเรยี ทางเดินอาหารส่วนกลาง ยุงลายบา้ น ยุงราคาญ ประเทศไทย 1กลุ่มการวจิ ยั โรคตดิ เช้อื ในสัตวท์ ่ีมีพาหะนาโรค หนว่ ยปรสติ วทิ ยา ภาควิชาพยาธิวิทยา คณะสัตวแพทยศาสตร์ จฬุ าลงกรณ์มหาวทิ ยาลยั 2ภาควิชาจุลชวี วิทยา คณะสัตวแพทยศาสตร์ จฬุ าลงกรณ์มหาวิทยาลัย 3หลกั สูตรสหสาขาวิชาชีวเวชศาสตร์ บณั ฑิตวทิ ยาลัย จฬุ าลงกรณ์มหาวทิ ยาลยั *ผูร้ บั ผิดชอบบทความ E-mail: [email protected], [email protected]

Original Article Plasmid profile of Enterococcus faecium and Enterococcus faecalis isolated from pigs, pork and humans in Thai-Laos border provinces Wink Phyo Thu1 Nuananong Sinwat2 Rungtip Chuanchuen3* Abstract This study aimed to determine plasmid profile of E. faecium and E. faecalis isolated from pigs, pig products and humans in Thailand-Laos border provinces. A total of 96 Enterococcus spp. including E. faecium (n=59) and E. faecalis (n=37) from pigs, pork and humans were included. All were determined for their antimicrobial susceptibilities and plasmid profile. Most isolates (96.9%) carried one to four plasmids with a molecular weight of 0.03-35 kb in E. faecium and 19-40 kb in E. faecalis. The 26 kb plasmid was most commonly detected. Variable plasmid profiles were defined, of which the most common pattern was, 23 kb in E. faecalis and, 35 kb in E. faecium. In conclusion, the results indicate a wide distribution of plasmids among the Enterococcus isolates originating from pig, pork and humans in this strain collection and confirm the important role of commensal E. faecium and E. faecalis as a reservoir for the emergence and spread of AMR among food animals and humans. Keywords: enterococcus, pig, plasmid, pork, Thai-Lao border provinces 1Department of Veterinary Public Health, University of Veterinary Science, Yezin, Nay Pyi Taw, 08252, Myanmar 2Department of Farm resources and production medicine, Faculty of Veterinary Medicine, Kasetsart University, Kamphaengsean campus, Nakhonpathom, 73140 Thailand 3Research Unit in Microbial Food Safety and Antimicrobial Resistance, Faculty of Veterinary Science, Chulalongkorn University, Bangkok 10330 Thailand *Correspondence: [email protected] Thai J Vet Med. 2018. 48(4): 623-630.

624 Thu W. P. et al. / Thai J Vet Med. 2018. 48(4): 623-630. Introduction (Vientiane and Savanakhate). All the isolates were stored in our strain collection. Enterococci are common microflora in the gastrointestinal tracts of humans and animals and are Enterococci were isolated as previously generally found in soil, water and plants. described (Domig et al., 2003). The species of The Enterococcus species is a leading cause of enterococci was identified by using multiplex PCR, of nosocomial infection in intensive care units and is which the primer sets used were FM1, frequently responsible for morbidity and mortality in GAAAAACAATAGAAGAATTAT and FM2, predisposed humans in hospitals (Zhang et al., 2017). TGCTTTTTTGAATTCTTCTTTA for the faecium Enterococci are intrinsically resistant to several antibiotics and capable of acquiring and transferring species and FL1, ACTTATGTGACTAACTTAACC and resistance elements (Murray, 1990), resulting in limited FL2, TAATGGTGAATCTTGGTTTGG were used for options for treatment of enterococcal infections (Arias the faecalis species (Jackson et al., 2004). A DNA and Murray, 2012). In addition, the bacteria have the ability to adapt to changing environmental conditions, template for PCR reaction was prepared using the leading to the particular challenge of elimination whole cell boiled lysate procedure as previously (Byappanahalli et al., 2012) and control of described (Levesque et al., 1995). E. faecalis ATCC antimicrobial resistance. 29212 was used as the positive control for E. faecalis. A representative PCR amplicon of E. faecium was Horizontal transfer of R plasmid among bacterial species is a key contributor to the spreading submitted for nucleotide sequencing and served as the of antimicrobial resistance (AMR) (Jensen et al., 2010). species control. One colony of each Enterococcus species Several resistant genes are plasmid mediated and antimicrobials have created selection pressure for the was collected from each positive sample. Research genes, resulting in persistence and the circulation of protocols involving human subjects were approved by resistance plasmids in bacteria. Even though the Ethics Committee of the Faculty of Medicine of enterococci of food animals and food-borne origin Khon Kaen University (The authorization ID, have not been identified as direct causes of clinical HE572136). infections, they are considered indirect causes of AMR and play a role as a reservoir of resistance determinants All PCR-species confirmed enterococci were that can be transferred to host-adapted strains (Thal et examined for their susceptibilities to 7 antimicrobials al., 1995). by determination of MICs using the two-fold agar dilution method (CLSI, 2008). The antimicrobials and Pigs serve as one of important sources of food their clinical MIC breakpoints from CLSI include in world communities and play a role as a major ampicillin (AMP, 16 µg/ml), chloramphenicol (CHL, reservoir for many species of bacteria, including 32 µg/ml), erythromycin (ERY, 8µg/ml), tetracycline enterococci. Antimicrobials have been widely used in (TET, 16 µg/ml) and vancomycin (VAN, 32 µg/ml). pig production for a long time (Barton, 2014). At the The clinical breakpoints from the National Thailand-Lao border, there are several crossing points Antimicrobial Resistance Monitoring System where cross border trade has taken place (Supatn, (NARMS) were used for gentamicin (GEN, 500 µg/ml) 2012). Pigs and pig products are among the common and streptomycin (STP, 1000 µg/ml) (NARMS, 2015). commodities and may be traded in a legal or illegal E coli ATCC 25922, Staphylococcus aureus ATCC 29213 manner (FAO et al., 2009). Together with the high and E. faecalis ATCC 29212 served as quality control frequency of border-crossing of humans and animals, trade of pigs and their meat may contribute to the strains. Multidrug resistance was defined as being distribution of AMR bacteria and their resistance resistant to three or more antimicrobials of different determinants. Up to date, there is very limited classes (Magiorakos et al., 2012). knowledge on the plasmid profile of E. faecium and E. faecalis along the food chain, particularly in the region. Plasmids were isolated from all the Therefore, this study was conducted to determine the Enterococcus isolates using the alkaline lysis method plasmid profile of E. faecium and E. faecalis isolated from pigs, pig products and humans in Thailand-Laos with some modifications (Jackson et al., 2012). Briefly, border provinces. a pellet of 1.5 ml overnight bacterial culture in Brain Heart Infusion broth was collected by centrifugation, Materials and Methods re-suspended in 100 µl TE buffer containing sucrose (10 mMTris, 1 mM EDTA, 25% sucrose, pH 8) and 1 All chemicals used in this study were A total mg/ml lysozyme (BiobasicInc®, Markham, Canada) of 96 Enterococcus isolates (i.e. E. faecium, n=59 and E. and incubated for 1 hr at 37˚C. The bacterial cells were faecalis, n=37) were included in this study. They were then lysed at 37˚C for 30 mins in lysis solution (100 µl obtained from pigs (n=30) and pig carcasses (n=29) of 0.2M NaOH and 1%sodium dodecyl sulfate (SDS, from slaughterhouses; pork (n=22) from retail markets Vivantis®, Selangor Darul Ehsan, Malaysia). A and humans including workers from slaughterhouses hundred-fifty µl of 3 M potassium acetate, pH 4.8 was (n=4), butchers in retail markets (n=2) and hospitalized added. After incubation on ice for 15 minutes, 350 µl of patients (n=9) between September, 2013 and October, phenol: chloroform: isoamyl alcohol (25:24:1) was 2014.The strains were isolated from rectal swabs from added and the suspension was centrifuged at 16,000×g pigs and humans, and carcass swabs from pig carcasses and retail pork in the border provinces of for 5 minutes. The aqueous phase was transferred to a Thailand (Nong Kai and Mukdaharn) and Lao PDR fresh Eppendrof tube and plasmid DNA was precipitated in 750 µl cold absolute ethanol. The DNA pellet was collected by centrifuging at 16,000×g for 10 minutes and washing in 70% cold ethanol. Plasmid DNA was re-suspended in 50 µl of TE buffer (10 mMTris, 1 mM EDTA, pH 8). The purified plasmids were separated on 0.8% agarose gel electrophoresis and stored at -20˚C.

Thu W. P. et al. / Thai J Vet Med. 2018. 48(4): 623-630. 625 The standard curve was created by plotting antimicrobial (Table 1). None of the E. faecalis isolates log10 molecular weight (kb) of known DNA sizes was resistant to ampicillin. The E. faecium isolates were versus their migration distance (mm). The relative commonly resistant to erythromycin (74.6%), mobility on agarose gel was used to estimate molecular streptomycin (72.9%) and tetracycline (88.1%). Most of weight of unknown plasmids. Plasmid profiles were the E. faecalis isolates exhibited resistance to defined based on the size and number of plasmids in erythromycin (89.2%), gentamicin (86.5%), each strain. streptomycin (86.5%) and tetracycline (97.3%). Resistant E. faecium was predominant in pig carcasses Results (93.3%) followed by pig isolates (90.0%). For E. faecalis, the percentages of resistance were variable among the In this study, most of the E. faecium and E. isolates of different origins. faecalis isolates were resistant to at least one Table 1 AMR phenotype in E. faecium (n=59) and E. faecalis (n=37) isolated from pigs, pig carcasses, retail pork and humans in Thailand and Lao PDR border provinces Species (n) Sample type No. of the isolates (%) (n) AMP CHL ERY GEN STR TET Pig (20) 8(40.0) 5(25.0) 15(75.0) 1 (5.0) 16(80.0) 18(90.0) E. faecium Pig carcass 8(53.3) 4(26.7) 12(80.0) 2(13.3) 12(80.0) 14(93.3) (59) (15) 5(31.3) 4(25.0) 10(62.5) 2(12.5) 10(62.5) 14(87.5) 3(37.5) 1(12.5) 7(87.5) 2(25.0) 5(62.5) 6(75.0) Retail meat (16) Human (8) Total 24(40.7) 14(23.7) 44(74.6) 7(11.9) 43(72.9) 52(88.1) Pig (10) 0 5(50) 10(100) 10(100) 8(80) 10(100) E. faecalis Pig carcass 1(7.1) 5(35.7) 14(100) 12(85.7) 12(85.7) 13(92.9) (37) (14) 0 3(50.0) 5(83.3) 4(66.7) 6(100) 6(100) 0 2(28.6) 4(57.1) 6(85.7) 6(85.7) 7(100) Retail meat (6) Human (7) Total 1(2.7) 15(40.5) 33(89.2) 32(86.5) 32(86.5) 36(97.3) AMP, ampicillin; CHL, chloramphenicol; ERY, erythromycin; GEN, gentamicin; STR: streptomycin; TET, tetracycline. Multidrug resistance was observed in 67.8% found among the E. faeciumisolates with a different of E. faecium and 89.2% of E. faecalis isolates, of which 18 AMR patterns of E. faecium and 11 AMR patterns of AMR pattern. A pig isolate with AMP-STR-TET E. faecalis were identified (Table 2). The most common resistance pattern and a pig carcasses isolate with ERY- resistance patterns were ERY-STR-TET in E. faecium TET resistance pattren did not carry plasmid. (16.9%) and ERY-GEN-STR-TET in E. faecalis (40.5%). Plasmids commonly found in E. faecalis High-level gentamicin resistance (HLGR) and high-level streptomycin resistance (HLSR) is defined isolates of pigs, pig carcass, retail pork and humans by having MIC value of 500 µg/ml for gentamicin and were 19-28 kb, 19-34 kb, 24-34 kb and 24-40 kb is size, 2,000 µg/ml for streptomycin (Chow, 2000; Klare et al., respectively. A 23 kb plasmid was most commonly 2003). The HLGR and HLSR phenotype was found in detected. The most common resistance pattern both E. faecium (11.9% and 49.2%, respectively) and E. identified in plasmid-carrying isolates was ERY-GEN- faecalis (86.5% and 73.0%, respectively). STR-TET. Only one human isolate was plasmid free. Almost all E. faecium (96.6%) and E. faecalis (97.3%) carried plasmids. The size of plasmids ranged from 0.03 to 35 kb in E. faecium and from 19 to 40 kb in E. faecalis. The 26 kb plasmid was most commonly observed among the enterococci. Mostly, enterococci carried only one plasmid. Three E. faecium isolates with the resistance pattern of AMP-CHL-ERY-STR-TET, ERY-STR-TET and STR- TET from pigs harbored 2 to 4 plasmids. All E. faecalis carried only one plasmid. Among the E. faecium isolates,the common plasmids in the pig and pork isolates and the human isolates were 23-35 kb and 29-35 kb in size, respectively. A 35 kb plasmid was most commonly

626 Thu W. P. et al. / Thai J Vet Med. 2018. 48(4): 623-630. Table 2 Plasmid profile of E. faecium (n=59) and E. faecalis (n=37) isolated from pigs, pig carcasses, retail pork and humans in Thailand and Lao PDR border provinces Size of plasmid E. faecium (n=59) E. faecalis (n=37) 626 Thu W. P. et al. / Thai J Vet Med. 2018. 48(4): 623-630. (kb) 40 Pig Pig carcass Retail pork Human AMR pattern Pig Pig carcass Retail pork Human AMR pattern 35 (n=20) (n=15) (n=16) (n=8) - (n=10) (n=14) (n=6) (n=7) - - - 1 CHL-STR-TET 35, 0.6, 0.4 - - - - - - - 1 GEN-STR-TET 34 - 34, 27 -- - 1 AMP-CHL-ERY-GEN-STR-TET - - - - 32 -1 - - AMP-CHL-ERY-STR-TET -- 1 - CHL-ERY-STR-TET -- - -- 31 1- - - AMP-ERY-STR - CHL-ERY-GEN-STR-TET 30 -- 1 - AMP-STR-TET - -1 1 ERY-GEN-STR-TET 11 - - ERY-STR-TET - ERY-GEN 29 -- - 1 ERY-STR - -1 -- 1 - TET 1 CHL-ERY-GEN-STR-TET 28 - 1- 1- - - ERY-STR-TET -- 27 - -1 - CHL-ERY-GEN-STR-TET 26 -- - 1 AMP- ERY-GEN-STR-TET 1- - - ERY- TET -- - -- - 1 ERY - 1- 1- - - STR-TET -- - -- -- - 1 AMP-ERY-STR-TET 1- - 1 ERY-STR-TET 1 2- - ERY-GEN-STR-TET -- 1 - ERY-STR - -1 - ERY-TET -- - 1 ERY-TET -- 1 - TET - 1- - ERY-GEN-STR-TET 2-- - CHL-ERY-GEN-STR-TET -- 1 - AMP-TET 2-- - ERY-GEN-STR-TET -1 - - AMP-ERY-STR-TET 1- - - AMP-STR-TET -1 - - AMP-ERY-STR -1 - - AMP-ERY-TET 1- - - ERY-STR-TET 1- - - STR-TET -1 1 - AMP-ERY-TET -- 1 - ERY-STR-TET -- 1 - STR-TET -- 1 - ERY -- - 1 TET 1- - - AMP-CHL-ERY-GEN-STR-TET -1 - - AMP-STR-TET -- 1 - CHL-STR-TET -1 - - STR-TET 1- - - AMP-TET 11 1 - AMP-ERY-GEN-STR-TET -1 1 - CHL-ERY-STR-TET -1 - - AMP-CHL-ERY-GEN-STR-TET 21 - - CHL-ERY-STR-TET

Thu W. P. et al. / Thai J Vet Med. 2018. 48(4): 623-630. 627 1- 1 - ERY-STR-TET 1- Thu W. P. et al. / Thai J Vet Med. 2018. 48(4): 623-630. -- -1 - - STR-TET -- 1- - - ERY-TET 1- -- 25 1- - - AMP-ERY-STR-TET - -1 - ERY-GEN-STR-TET -- - ERY-GEN-TET -1 - - ERY-STR-TET 1 -- 1- -- -- 1 - ERY-TET -- - CHL-ERY-GEN-STR-TET 2- 1 ERY-GEN-STR-TET 25, 0.3, 0.1, 0.03 1 - - - AMP-CHL-ERY-STR-TET - 1- - CHL-ERY- STR-TET 1- 1 GEN-STR-TET 24 1- - - CHL-ERY-TET - 1- - CHL-ERY-GEN-STR-TET 1- - AMP-ERY-STR-TET - -- - CHL-ERY-GEN-TET -1 - ERY-GEN-STR-TET - -- - CHL-ERY-TET - CHL-ERY-GEN-STR-TET - - CHL-ERY-GEN-STR-TET - ERY-GEN-STR-TET 23 -- 1 - AMP-CHL-ERY-GEN-STR-TET 1 - CHL-ERY-GEN-STR-TET - ERY-GEN-STR-TET - 1 ERY-GEN-TET 1 - - 22 -- - -- - 21 -- - -- - 20 -- - -- 1 19 -- - -- 1 - No plasmid 1 - - - AMP-STR-TET - detected -1 - - ERY-TET -, not found AMP, ampicillin; CHL, chloramphenicol; ERY, erythromycin; GEN, gentamicin; STR, streptomycin; TET, tetracycline. 627

628 Thu W. P. et al. / Thai J Vet Med. 2018. 48(4): 623-630. Discussion the Enterococcus isolates carried a single plasmid, only three E. faecium isolates harbored more than one As commensals, E. faecalis and E. faecium are plasmid (2-3 plasmids). The difference in size and part of target bacteria suggested to be included in the number of plasmids may be due to the plastic and AMR surveillance program in food animals (EFSA, dynamic structure of plasmids (Jensen et al., 2010). The 2012). The antimicrobials tested in this study were environment where the plasmids existed may be selected based on recommendation for the harmonized involved in the variation in the size and number of panel of antimicrobials for Enterococcus spp., in AMR plasmids (Jensen et al., 2010). Plasmids are unstable monitoring. The data reveals that enterococci were and their transfer can be influenced by environmental resistant to at least one antimicrobial excluded changes and the plasmid profile of enterococci was not vancomycin. More than half of E. faecium (67.8%) and identical although the cultures were prepared on the E. faecalis (89.2%) were multidrug resistant bacteria. same day (Jackson et al., 2012). Therefore, plasmid Even though the number of isolates in this study was analysis should be performed in combination with limited, the data still demonstrated that most other molecular techniques. Enterococci were multidrug resistant strains. Vancomycin resistance in E. faecium is a particular In this study, most of E. faecium (69.5%) and E. health concern. In this study, none of the isolates faecalis (83.8%) isolates carrying plasmids were exhibited vancomycin resistance, in agreement with a resistant to aminoglycosides (i.e. gentamicin and previous study conducted in the swine isolates from streptomycin). This is in agreement with a previous Northern Thailand (Love et al., 2015). A linkage study reporting that aminoglycoside-modifying between avoparcin and VRE in animals was enzyme determinants in E. faecium and E. faecalis were demonstrated (Klare et al., 2003). Growth promoter use plasmid borne (Coleri et al., 2004). of avoparcin in livestock in Thailand has been banned since 1998 and the VRE prevalence in poultry The enterococcal isolates of pigs, pig products production has been gradually decreased and humans from Thailand and Lao PDR border (Matayompong, 2012). This could explain the absence provinces were resistant to multiple drugs, particularly of vancomycin resistance observed in this study. In gentamicin, erythromycin and tetracycline. The addition, the most common AMR patterns were ERY- enterococci are commensal microflora, therefore, such STR-TET and ERY-GEN-STR-TET in E. faecium (16.9%) distribution of AMR implies the wide use of these and E. faecalis (40.5%), respectively. This is consistent antimicrobials in pig production and human medicine with a previous report in Denmark (Aarestrup et al., in the countries. It additionally indicates the horizontal 2000). transfer and widespread nature of resistance determinants locating on the same plasmids. Plasmids are extrachromosomal DNA carrying antimicrobial resistance determinants that can In conclusion, the results from this study be acquired and deleted. Their compositions can confirms the important role of commensal E. faecium change rapidly and the epidemiologically related and E. faecalis as a reservoir for the emergence and isolates may exhibit different plasmid profiles (Jensen spread of AMR among food animals. Data on the et al., 2010). Plasmids play a role as the reservoirs for burden and distribution of AMR involved in national the intra- and inter-species transmission of resistance and regional surveillance program is required for the determinants (Rowe-Magnus and Mazel, 2001; Giraffa, better control and prevention of AMR. Further studies, 2002) and horizontal transfer of resistance plasmids including molecular characterization of plasmids, are was previously demonstrated enterococci (Giraffa, warranted to enhance understanding of emergence 2002; Choi and Woo, 2015). Plasmid analysis may not and dissemination of resistant enterococci. be very useful in discriminating between epidemic and endemic strains but can benefit in describing a link of Acknowledgements AMR from different sources. This work was financially supported by a In this study, plasmids were frequently Research Grant for Mid-Career University Faculty isolated from either E. faecium (96.6%) or E. faecalis RSA5680051 confounded by the Thailand Research (97.3%). All the isolates in this collection harbored only Fund (TRF), Faculty of Veterinary Science, one plasmid, with the exception of three E. faecium Chulalongkorn University. It was partly supported by isolates that contained two to four plasmids. A 90th Anniversary of Chulalongkorn University Fund previous study explained that E. faecium and E. faecalis (Ratchadaphiseksomphot Endowment Fund, Batch 32, isolates from naturally fermented foods carried one to 3/2016). WPT is a recipient of the ASEAN scholarship six plasmids (Togay et al., 2010). Another study of program for ASEAN countries, Chulalongkorn clinical E. faecalis isolates carried one to five plasmids University. (Song et al., 2013). It should be noted that the copy number of the isolated plasmid may be affected by the References plasmid isolation technique. Large denatured plasmids cannot renature as fast as small ones and may Arias CA and Murray BE 2012. The rise of the be lost during the DNA precipitation step. Enterococcus: beyond vancomycin resistance. Nat Rev Microbiol. 10(4): 266-278. Plasmids in variable size were detected among the isolates in this study. The 35 kb and 23 kb Arias CA and Murray BE 2012. The rise of the plasmids were most commonly detected in E. faecium Enterococcus: beyond vancomycin resistance. Nat and E. faecalis, in agreement with previous studies Rev Microbiol. 10(4): 266-278. (Togay et al., 2010; Barua et al., 2016). While almost all Barton MD 2014. Impact of antibiotic use in the swine industry. Curr Opin Microbiol. 199-15.

Thu W. P. et al. / Thai J Vet Med. 2018. 48(4): 623-630. 629 Barua M, Das S, Gupta C, Saha R and Kaur IR 2016. Love DC, Tharavichitkul P, Arjkumpa O, Imanishi M, Plasmid profile and antibiogram of Enterococcal faecalis isolated from tertiary care hospital in Hinjoy S, Nelson K and Nachman KE 2015. Delhi. Indian J Med Microbiol. 34(1): 109-110. Antimicrobial Use and Multidrug-Resistant Byappanahalli MN, Nevers MB, Korajkic A, Staley ZR Salmonella spp., Escherichia coli, and Enterococcus and Harwood VJ 2012. Enterococci in the environment. Microbiol Mol Biol Rev. 76(4): 685- faecalis in Swine from Northern Thailand. Thai J 706. Choi JM and Woo GJ 2015. Transfer of tetracycline Vet Med. 45(1): 43-53. resistance genes with aggregation substance in food-borne Enterococcus faecalis. Curr Microbiol. Magiorakos AP, Srinivasan A, Carey RB, Carmeli Y, 70(4): 476-484. Falagas ME, Giske CG, Harbarth S, Hindler JF, Chow JW 2000. Aminoglycoside resistance in Kahlmeter G, Olsson-Liljequist B, Paterson DL, enterococci. Clin Infect Dis. 31(2): 586-589. CLSI 2008. Performance standards for antimicrobial Rice LB, Stelling J, Struelens MJ, Vatopoulos A, disk and dilution susceptibility tests for bacteria Weber JT and Monnet DL 2012. Multidrug- isolated from animals; Approved standard- Third edition M31-A3: Wayne, PA, USA. pp. resistant, extensively drug-resistant and Coleri A, Cokmus C, Ozcan B, Akcelik M and Tukel C 2004. Determination of antibiotic resistance and pandrug-resistant bacteria: an international resistance plasmids of clinical Enterococcus expert proposal for interim standard definitions species. J Gen Appl Microbiol. 50(4): 213-219. Domig KJ, Mayer HK and Kneifel W 2003. Methods for acquired resistance. Clin Microbiol Infect. used for the isolation, enumeration, 18(3): 268-281. characterisation and identification of Enterococcus Matayompong P 2012. Vancomycin Resistant spp. 1. Media for isolation and enumeration. Int J Food Microbiol. 88(2-3): 147-164. Enterococci -Thailand Experience. International EFSA 2012. Technical specifications on the harmonised monitoring and reporting of antimicrobial Symposium on “Recent Progress in Swine resistance in Salmonella, Campylobacter and indicator Escherichia coli and Enterococcus spp. Breeding and Raising Technologies”. bacteria transmitted through food. The EFSA http://cdn.aphca.org/dmdocuments/AMR%20 Journal. 10(6): 1-64. FAO, ADB, OIE and SEAFMD. 2009. \"Bangkok, WS%20Proceedings_121112.pdf (accessed Thailand: Study on cross-border movement and market chains of large ruminants and pigs in the 7.6.2016). Greater Mekong SubRegion.\" [Online]. Available: http://ulm.animalhealthresearch.asia/newslette Murray BE 1990. The life and times of the Enterococcus. rs/FAO_ADB_OIE_Cross- Border%20movement%20study_Final%20Report Clin Microbiol Rev. 3(1): 46-65. .pdf. Accessed June 6, 2016. Giraffa G 2002. Enterococci from foods. FEMS NARMS. 2015. \"Data Tables in 2012-2013 Integrated Microbiol Rev. 26(2): 163-171. Jackson CR, Fedorka-Cray PJ and Barrett JB 2004. Use NARMS Report. FDA.\" [Online]. Available: of a genus- and species-specific multiplex PCR for identification of enterococci. J Clin Microbiol. http://www.fda.gov/AnimalVeterinary/Safety 42(8): 3558-3565. Jackson CR, Spicer LM, Barrett JB and Hiott LM 2012. Health/AntimicrobialResistance/NationalAnti Application of Multiplex PCR, Pulsed-Field Gel Electrophoresis (PFGE), and BOX-PCR for microbialResistanceMonitoringSystem/ucm0591 Molecular Analysis of Enterococci. Gel Electrophoresis - Principles and Basics.269-298. 03.htm. Accessed June 7, 2016. Jensen LB, Garcia-Migura L, Valenzuela AJ, Lohr M, Hasman H and Aarestrup FM 2010. A Rowe-Magnus DA and Mazel D 2001. Integrons: classification system for plasmids from enterococci and other Gram-positive bacteria. J natural tools for bacterial genome evolution. Curr Microbiol Methods. 80(1): 25-43. Klare I, Konstabel C, Badstubner D, Werner G and Opin Microbiol. 4(5): 565-569. Witte W 2003. Occurrence and spread of antibiotic resistances in Enterococcus faecium. Int J Song X, Sun J, Mikalsen T, Roberts AP and Sundsfjord Food Microbiol. 88(2-3): 269-290. A 2013. Characterisation of the plasmidome Levesque C, Piche L, Larose C and Roy PH 1995. PCR within Enterococcus faecalis isolated from mapping of integrons reveals several novel combinations of resistance genes. Antimicrob marginal periodontitis patients in Norway. PLoS Agents Chemother. 39(1): 185-191. One. 8(4): e62248. Supatn N 2012. \"A study on cross-border trade facilitation and regional development along economic corridors: Thailand perspectives\" in emerging economic corridors in the Mekong Region. Bangkok Research Report No. 8.(IDE- JETRO): 232-270. Thal LA, Chow JW, Mahayni R, Bonilla H, Perri MB, Donabedian SA, Silverman J, Taber S and Zervos MJ 1995. Characterization of antimicrobial resistance in enterococci of animal origin. Antimicrob Agents Chemother. 39(9): 2112-2115. Togay SO, Keskin AC, Acik L and Temiz A 2010. Virulence genes, antibiotic resistance and plasmid profiles of Enterococcus faecalis and Enterococcus faecium from naturally fermented Turkish foods. J Appl Microbiol. 109(3): 1084- 1092. Zhang Y, Du M, Chang Y, Chen LA and Zhang Q 2017. Incidence, clinical characteristics, and outcomes of nosocomial Enterococcus spp. bloodstream infections in a tertiary-care hospital in Beijing, China: a four-year retrospective study. Antimicrob Resist Infect Control. 6(73):1-11.

63280 Thu W. P. et al. / Thai J Vet Med. 2018. 48(4): 623-630. 1 บทคัดยอ่ รปู แบบพลาสมิดของ Enterococcus faecium และ Enterococcus faecalis ที่แยกไดจ้ ากสุกรเน้อื สุกรและคน Wink Phyo Thu1 นวลอนงค์ สินวตั 2 รงุ่ ทิพย์ ชวนช่ืน3* การวิจยั คร้ังน้มี วี ตั ถุประสงคเ์ พ่ือศกึ ษารปู แบบพลาสมิดของ Enterococcus faecium และ E. faecalis ที่แยกไดจ้ ากสกุ ร เนื้อสกุ ร และคนในจังหวดั ชายแดนประเทศไทยและลาว E. faecium (n=59) และ E. faecalis (n=37) จานวน 96 isolates แยกได้จากสกุ ร เนอื้ สกุ ร และคน นามาทดสอบความไวตอ่ ยาตา้ นจุลชีพและศกึ ษารูปแบบพลาสมิด พบวา่ เชื้อส่วนใหญ่ (96.9%) มีพลาสมิดจานวน 1-4 พลาสมดิ พบพ ลาสมิดขนาด 26 kb มากทส่ี ุด รูปแบบพลาสมิดมคี วามหลากหลาย โดยท่พี บมากทสี่ ดุ ใน E. faecalis และE. faecium คอื พลาสมิดท่มี ขี นาด 23 kb และ และ 35 kb ตามลาดบั ผลการวิจยั ชใี้ หเ้ หน็ ถงึ การกระจายตัวอย่างกว้างขวางของพลาสมดิ ใน Enterococcusท่ีแยกได้จากสุกร เนือ้ สกุ รและคนในจงั หวดั ชายแดนประเทศไทยและลาว ยืนยันบทบาทสาคญั ของ E. faecalis และ E. faecium ท่ีเปน็ แบคทเี รยี ประจาถน่ิ ต่อ การอุบตั ิและแพร่กระจายของการดอ้ื ยาในสกุ รและคน คาสาคัญ: เอน็ เทอโรคอคคัส สกุ ร รปู แบบพลาสมิด เน้อื สกุ ร จงั หวดั ชายแดนประเทศไทยและลาว 1Department of Veterinary Public Health, University of Veterinary Science, Yezin, Nay Pyi Taw, 08252, Myanmar 2ภาควชิ าเวชศาสตรแ์ ละทรัพยากรการผลติ สัตว์ คณะสตั วแพทยศาสตรม์ หาวทิ ยาลัยเกษตรศาสตร์ วทิ ยาเขตกาแพงแสน นครปฐม 73140 3หนว่ ยปฏิบัตกิ ารความปลอดภัยอาหารทางจลุ ชวี วทิ ยาและเชอ้ื ด้ือยา ภาควิชาสัตวแพทยสาธารณสขุ คณะสัตวแพทยศาสตร์ จุฬาลงกรณ์ มหาวิทยาลัย กรงุ เทพฯ 10330 *ผู้รบั ผดิ ชอบบทความ E-mail: [email protected]

Original Article Oxidative stress and antioxidant in canine cutaneous mast cell tumors Arayaporn Macotpet1* Ekkachai Pattarapanwichien2 Fanan Suksawat1 Patcharee Boonsiri3 Abstract Oxidative stress can result from either the excessive production of reactive oxygen species (ROS) or an impaired antioxidant system, or both. It causes damage to lipids, proteins and DNA. Therefore, oxidative stress may be involved in carcinogenesis, and is associated with many types of cancer in dogs. The objective of this study was to compare the levels of malondialdehyde, protein hydroperoxides, glutathione, retinol and alpha-tocopherol between dogs with mast cell tumors and clinically healthy controls. Blood samples were obtained from eighteen clinically healthy dogs and fourteen dogs with spontaneous mast cell tumors. Malondialdehyde and protein hydroperoxides levels were measured by the thiobarbituric acid reactive substance assay, and the ferric-xylenol orange assay, respectively. Glutathione level was determined using spectrophotometric assay. Retinol and alpha-tocopherol levels were measured using the high performance liquid chromatographic method. Dogs with mast cell tumors had significantly higher levels of malondialdehyde (P<0.01) and protein hydroperoxides (P<0.05) compared with the clinically healthy controls. When considering antioxidants, dogs with mast cell tumors had significantly lower levels of glutathione (P<0.01), retinol (P<0.05) and alpha-tocopherol (P<0.01) compared with the clinically healthy controls. Mast cell tumors in dogs are associated with oxidative stress and antioxidant status. Further studies on oxidative stress and antioxidant activity in dogs should be conducted to guide and plan the complementary treatment of canine cancer. Keywords: dogs, mast cell tumors, malondialdehyde, protein hydroperoxides, glutathione, retinol, alpha-tocopherol 1Division of Small Animal Medicine, Faculty of Veterinary Medicine, Khon Kaen University, Khon Kaen, 40002, Thailand 2Division of Pathobiology, Faculty of Veterinary Medicine, Khon Kaen University, Khon Kaen, 40002, Thailand 3Department of Biochemistry, Faculty of Medicine, Khon Kaen University, Khon Kaen, 40002, Thailand *Correspondence: [email protected] Thai J Vet Med. 2018. 48(4): 631-637.

632 Macotpet A. et al. / Thai J Vet Med. 2018. 48(4): 631-637. Introduction Hospital, Faculty of Veterinary Medicine, Khon Kaen University, Khon Kaen, Thailand. Of 65 dogs, 19 dogs Mast cell tumors (MCTs) are one of the most were identified with grade II MCTs and 46 dogs were common malignant cutaneous tumors in dogs, and detected as being clinically healthy. Five of the dogs often demonstrate aggressive biological behavior (Jark with grade II MCTs and 28 clinically healthy dogs were et al., 2017). However, they can be found in excluded from the study, as they did not achieve the extracutaneous locations, such as the oral cavity, inclusion criteria. Therefore, 14 dogs with grade II nasopharynx, salivary gland, etc. (Patnaik et al., 1982; MCTs and 18 clinically normal dogs were included in Carberry et al., 1987). MCTs may spread to nearby the present study. lymph nodes, spleen, liver, other skin sites and distant organs. MCTs can be classified into 3 grades. Grade I There were 18 clinically healthy dogs: 7 males MCTs develop slowly and persist for years without and 11 females with an average age of 8.17 years (5-12 increasing in size, whereas grade III MCTs show years old), and 14 dogs with grade II MCTs: 7 males aggressive growth and have a high recurrence and 7 females with an average age of 8.87 years (5-12 potential. Grade II MCTs are intermediate between the years old). For inclusion criteria, grade II MCTs- other two forms. More than 80% of grade III MCTs bearing dogs had to be diagnosed by a histopathology metastasize and frequently cause death (Welle et al., test according to the classical Patnaik grading system 2008). The prognosis for grade II MCTs is variable and (Patnaik et al., 1984) and were classified as stage III they may cause death in 17-56% of cases due to the disease (The World Health Organization (WHO) failure of treatment or metastatic status (Blackwood et classification scheme for canine mast cell tumors). al., 2012). Dogs receiving chemotherapy or antioxidant supplements were excluded from the study. Clinically There is much evidence that reactive oxygen healthy dogs had to be aged 5-12 years old without species (ROS) are involved in the aetiopathogenesis of blood parasites or intestinal parasites and they had to many diseases such as cancer (Waris and Ahsan, 2006; have had a normal physical examination at least six Choudhari et al., 2014; Katakwar et al., 2016). ROS can months before the blood collection. Owner consent seriously alter the structure of biomolecules such as was obtained for all participating dogs, and the study lipids, proteins and deoxyribonucleic acid (DNA). The protocol was approved by the Animal Ethics imbalance between the rates of ROS production and Committee of Khon Kaen University. ROS removal is known as oxidative stress, and this is involved in carcinogenesis in dogs (Macotpet et al., Blood sample collection and preparation: Four 2013). Malondialdehyde (MDA) is one of the lipid milliliters of blood was drained from the cephalic vein peroxidation products commonly used as a biomarker and separated into 2 parts. One ml was treated with of oxidative stress (Gawel et al., 2004). Oxidative ethylenediamine tetraacetic acid (EDTA), for a damage of proteins yields protein hydroperoxides, complete blood count analysis and the presence of which are occasionally utilized for the evaluation of blood parasites. The remainder was centrifuged at 3000 oxidative stress (Gebicki, 1997). rpm at 4°C for 10 min to provide serum for blood chemistry tests. Antioxidants act as “free radical scavengers” by preventing and repairing damages caused by ROS, Complete blood count and blood chemistry analysis: and can improve immune protection and lower the risk Packed cell volume (PCV), hemoglobin (Hb), red blood of cancer and other diseases (Valko et al., 2006). cell counts (RBC) and white blood cell counts (WBC) Glutathione (GSH) is an antioxidant which participates were measured using an automatic analyzer (XT- in the body’s defense against free radicals and 2000iV: Sysmex®, USA). Creatinine and alanine carcinogens. A retrospective study of natural and aminotransferase (ALT) were assayed using an synthetic retinoids has demonstrated an inhibition of automatic blood chemistry analyzer (AU400; the growth and the development of different types of Olympus®, USA). cancers in human and dogs (Hong et al., 2000b; Ohashi et al., 2001; Pinello et al., 2009; Tang and Gudas, 2011). Determination of oxidative stress marker and Alpha-tocopherol is the most naturally occurring antioxidant levels: Oxidative stress markers (MDA compound of vitamin E and is an effective antioxidant and protein hydroperoxides) and antioxidant (retinol, that prevents the incidence of some cancers by alpha-tocopherol and GSH) were determined using protecting cells and DNA from free radicals (Traber serum samples preserved at -80°C. MDA and protein and Atkinson, 2007; Rizvi et al., 2014). hydroperoxides were measured by the thiobarbituric acid reactive substance assay (TBARS) (Nielsen et al., The purpose of this study was to compare 1997) and the ferric–xylenol orange peroxide assay some biomarkers of oxidative stress (MDA and protein (Gay et al., 1999), respectively. GSH level was hydroperoxides) and antioxidant (glutathione, retinol determined using 5-5’dithiobis [2-nitrobenzoic acid] and alpha-tocopherol) levels between dogs with grade (DTNB) reagent (Shigesawa et al., 1992). Retinol and II MCTs, and clinically healthy controls. The alpha-tocopherol, were measured using a high knowledge from this study may introduce a performance liquid chromatographic (HPLC) method complementary method, such as the supplementation modified from Thurnham (Thurnham et al., 1988). of food rich in antioxidants, for the prevention or treatment of Grade II MCTs in dogs. Statistical analysis: Data is presented as mean ± SD. Parametric data was determined using the Materials and Methods independent-sample t test whereas the Mann-Whitney Animals: Dogs were registered in this study between July-November 2017 at the Veterinary Teaching

Macotpet A. et al. / Thai J Vet Med. 2018. 48(4): 631-637. 633 U test was used for non-parametric data. All statistical ranges; however, red blood cell counts were analyses were performed by statistical software (SPSS significantly lower in dogs with grade II MCTs than in version 17; SPSS Inc., Chicago, Ill.) and values of P<0.05 clinically healthy dogs (P<0.001). No significant were considered significant. differences in white blood cell counts between clinically healthy and grade II MCTs groups (P=0.28) Results were found. Notably, an upward trend in white blood cell counts was observed in dogs with grade II MCTs. There was no significant difference in age and Creatinine and ALT levels were within normal ranges body weight between groups (P=0.36 and 0.42, and not significantly different between the two groups respectively). The proportion of pure breeds in the (P=0.054 and P=0.14, respectively). dogs with grade II MCTs and the clinically healthy dogs showed no difference between groups (5/14 or The oxidative stress marker levels show that 35.71% vs 6/18 or 33.34%, respectively). In the dogs serum MDA was significantly higher in dogs with with grade II MCTs, the pure breeds were Rottweiler grade II MCTs than in clinically healthy dogs (Table 1). (n=1), Thai Ridgeback (n=2) and Golden Retriever Similarly, serum protein hydroperoxides were (n=2). In the clinically healthy dogs, the pure breeds significantly higher in dogs with grade II MCTs than in were Golden Retriever (N=2), Labrador Retriever clinically healthy dogs. The results in Table 1 also show (n=2), Poodle (n=1) and Rottweiler (n=1). that antioxidant levels including serum retinol, alpha- tocopherol and GSH were significantly lower in dogs The numbers of red blood cells and white with grade II MCTs than in clinically healthy dogs. blood cells in both groups remained within normal Table 1 MDA, protein hydroperoxides, GSH, retinol and alpha-tocopherol for dogs with grade II mast cell tumors and clinically healthy dogs. Parameters dogs with grade II mast cell clinically healthy P-valuea dogs tumors <0.01 (N = 18) <0.05 (N = 14) 4.30 ± 0.49 <0.01 <0.05 MDA, 6.14 ± 1.67 8.64 ± 3.14 <0.01 mean ± SD, µmol/L 12.24 ± 1.39 protein hydroperoxides, mean ± 63.84 ± 62.52 3.68 ± 1.10 SD, µmol/L 19.33 ± 6.94 GSH, 6.73 ± 1.71 mean ± SD, mg/dL retinol, 2.58 ± 1.81 mean ± SD, µmol/L alpha-tocopherol, 10.64 ± 5.28 mean ± SD, µmol/L Abbreviations: GSH: glutathione; MDA: malondialdehyde Discussion oxidative stress marker levels as shown in Table 1, therefore, reflect that lipid and protein damage occur The imbalance between the rates of reactive in dogs with grade II MCTs. oxygen species (ROS) production and removal is known as oxidative stress, which leads to the The physiological level of ROS is usually degradation of lipids, proteins and nucleic acids. Lipid regulated by antioxidant defense mechanisms. GSH peroxidation products, including MDA, are often used acts as a powerful antioxidant by working as biomarkers for oxidative stress status (Macotpet et synergistically with the other antioxidants to al., 2013). In the present study, serum MDA levels in neutralize and scavenge ROS and prevent oxidative dogs with grade II MCTs were significantly higher stress (Li et al., 2004; Kerksick and Willoughby, 2005). than those of the control group (Table 1). This implies In this study, the GSH levels were significantly lower that the formation of free radicals and consequent lipid in the grade II MCTs group than in the clinically peroxidation may be related to mast cell tumors. healthy group (P<0.001) (Table 1). Our results agree Moreover, protein hydroperoxides, major products of with a previous study (Kumar et al., 1995), which hydroxyl and singlet oxygen attack on the body reported a significantly lower level of GSH in grade III proteins during the burden of oxidative stress, are also cervical intraepithelial neoplasia and invasive cancer found in those with oxidative stress (Gebicki, 1997). compared to the controls. Furthermore, reduced GSH These hydroperoxides can cause oxidative damage and levels of the normal tissues were found to be higher may contribute to the changing of cellular redox than those in the tumoral tissues which may be a signaling and reducing antioxidants in the body consequence of increased detoxification activity in the (Davies, 2005; Gracanin and Davies, 2007). The results tumor cells (Saydam et al., 1997). The quantitative in Table 1 show significantly higher serum protein changes occurring in serum GSH is a useful finding hydroperoxides in dogs with mast cell tumor than in and it might represent a systemic biochemical marker the control group. In this study, the standard deviation for MCTs. of protein hydroperoxides in the cancerous group is quite high, and this may result from the extremely high Nowadays, antioxidant vitamins are being protein hydroperoxide level in 2 dogs. This may be studied for use in the treatment and prevention of explained by noting that hydroxyl and singlet oxygen diseases related to oxidative stress, including cancer. in these 2 dogs was high (Gebicki, 1997). The results of Most studies of antioxidant vitamins are conducted in humans rather than in animals. In this study, dog

634 Macotpet A. et al. / Thai J Vet Med. 2018. 48(4): 631-637. serum retinol and alpha-tocopherol, which are indicated a greater utilization of the vitamin. However, commonly known as vitamin A and E respectively, there has been a report that alpha-tocopherol did not were determined by HPLC. At the present time, the differ between healthy and dogs undergoing results of a study on vitamin A and its provitamin are chemotherapy for malignant neoplasms (Galler et al., still being debated between researchers about the 2015). effectiveness of antioxidants (Lewicka et al., 2017). In contrast, the effectiveness of antioxidants for vitamin E Conclusion: MCTs in dogs may be enhanced or are irrefutably confirmed by several studies (Traber developed if there is an imbalance between oxidative and Atkinson, 2007). stress and antioxidant status. The dogs with grade II MCTs in this study had increased oxidative stress Retinol, a derivative form of vitamin A, has (MDA and protein hydroperoxides) and decreased cytotoxic effects and the induction of apoptosis against antioxidants (GSH, retinol, and alpha-tocopherol). cancer (Bushue and Wan, 2010). Retinoic acid, a Further studies on oxidative stress and antioxidant derivative of retinol, can inhibit tumor growth by activity in dogs should be conducted to obtain clinical tumor suppressor genes regulation (Connolly et al., guidelines for the prevention or treatment of canine 2013) and it has been used successfully for the cancer. treatment and chemoprevention of acute promyelocytic leukemia and solid cancers (Hansen et Acknowledgements al., 2000; Lengfelder et al., 2000; Lengfelder et al., 2005). Retinoic acid, in combination with surgery, can extend This work was supported partially by a grant the survival rate of thyroid carcinoma and decrease the from the Faculty of Veterinary Medicine, Khon Kaen time to tumor recurrence (Castillo et al., 2016). In dogs, University, Khon Kaen, Thailand. We thank Mr.Ian retinoids encourage morphological differentiation and Thomas for editing the manuscript. growth inhibition in cell lines of canine osteosarcoma. Thus, retinoids may have the ability to act as an References adjunctive treatment for osteosarcoma and MCTs in dogs (Hong et al., 2000a; Ohashi et al., 2006). In vitro, Blackwood L, Murphy S, Buracco P, De Vos JP, De retinoic acid may be a potential chemotherapeutic Fornel-Thibaud P, Hirschberger J, Kessler M, agent for the treatment of canine MCTs (Pinello et al., Pastor J, Ponce F, Savary-Bataille K and Argyle DJ 2009). However, retinoids do not induce 2012. European consensus document on mast cell differentiation, apoptosis or growth inhibition of tumours in dogs and cats. Vet Comp Oncol. 10(3): melanoma cell lines in dog (Ohashi et al., 2001). e1-e29. In this study, retinol levels were significantly Bushue N and Wan YJ 2010. Retinoid pathway and lower in dogs with grade II MCTs than in clinically cancer therapeutics. Adv Drug Deliv Rev. 62(13): healthy dogs (Table 1). The possible explanation is that 1285-1298. these dogs had diets with low retinol. A similar study reported that dogs with low vitamin A in their meals Carberry CA, Flanders JA, Anderson WI and Harvey had low blood vitamin A level (Schweigert and Bok, HJ 1987. Mast cell tumor in the mandibular 2000). Retinol supplementation for dogs with MCTs salivary gland in a dog. Cornell Vet. 77(4): 362- during a cancer protocol might reduce MTCs risk 366. factors. A further study on retinol supplementation in dogs with cancer to encourage longer-term survival, Castillo V, Pessina P, Hall P, Blatter MF, Miceli D, Arias including prevention of cancer in dogs is warranted. ES and Vidal P 2016. Post-surgical treatment of thyroid carcinoma in dogs with retinoic acid 9 cis Vitamin E may be advantageous in the improves patient outcome. Open Vet J. 6(1): 6-14. inhibition of carcinogenesis and cancer metastasis as well as in the enhancement of immunity against the Choudhari SK, Chaudhary M, Gadbail AR, Sharma A cancer (Ricciarelli et al., 2001). In humans, vitamin E and Tekade S 2014. Oxidative and antioxidative has been suggested to reduce the risk of cancer. In mechanisms in oral cancer and precancer: a addition, vitamin E supplementation in dog food can review. Oral Oncol. 50(1): 10-18. prevent lethal ventricular arrhythmias associated with ischemia and reperfusion (Sebbag et al., 1994). Alpha- Connolly RM, Nguyen NK and Sukumar S 2013. tocopherol, one form of vitamin E in the body, acts as Molecular pathways: current role and future a cellular antioxidant defense system (Rizvi et al., 2014) directions of the retinoic acid pathway in cancer and it may decrease oxidation during exercise and prevention and treatment. Clin Cancer Res. 19(7): improve performance or recovery in dogs (Scott et al., 1651-1659. 2001). Davies MJ 2005. The oxidative environment and In this study, alpha-tocopherol levels were protein damage. Biochim Biophys Acta. 1703(2): significantly lower in dogs with grade II MCTs than in 93-109. clinically healthy dogs (Table 1). The results are supported by a previous study that alpha-tocopherol Galler AT, Krammer-Lukas S, Holler U, Thalhammer levels decreased in dogs with lymphoma when JG, Zentek J and Willmann M 2015. Blood vitamin compared to those of the control dogs (Winter et al., levels in dogs with malignancies and the 2009). Dogs with mammary cancer had decreased influence of chemotherapy. Wien Tierärztl Monat alpha-tocopherol levels in their neoplastic tissues Vet Med Austria. 102: 144-154. compared with the normal tissue of an adjacent mammary gland (Karayannopoulou et al., 2013) which Gawel S, Wardas M, Niedworok E and Wardas P 2004. Malondialdehyde (MDA) as a lipid peroxidation marker. Wiadomosci lekarskie. 57(9-10): 453-455.

Macotpet A. et al. / Thai J Vet Med. 2018. 48(4): 631-637. 635 Gay C, Collins J and Gebicki JM 1999. Hydroperoxide vitro study. Int J Occup Med Environ Health. assay with the ferric-xylenol orange complex. 30(5): 695-703. Anal Biochem. 273(2): 149-155. Li Y, Wei G and Chen J 2004. Glutathione: a review on biotechnological production. Appl Microbiol Gebicki JM 1997. Protein hydroperoxides as new Biotechnol. 66(3): 233-242. reactive oxygen species. Redox Report. 3(2): 99- Macotpet A, Suksawat F, Sukon P, Pimpakdee K, 110. Pattarapanwichien E, Tangrassameeprasert R and Boonsiri P 2013. Oxidative stress in cancer- Gracanin M and Davies MJ 2007. Inhibition of protein bearing dogs assessed by measuring serum tyrosine phosphatases by amino acid, peptide, malondialdehyde. BMC Vet Res. 9: 101. and protein hydroperoxides: potential Nielsen F, Mikkelsen BB, Nielsen JB, Andersen HR and modulation of cell signaling by protein oxidation Grandjean P 1997. Plasma malondialdehyde as products. Free Radic Biol Med. 42(10): 1543-1551. biomarker for oxidative stress: reference interval and effects of life-style factors. Clin Chem. 43(7): Hansen LA, Sigman CC, Andreola F, Ross SA, Kelloff 1209-1214. GJ and De Luca LM 2000. Retinoids in Ohashi E, Hong SH, Takahashi T, Nakagawa T, chemoprevention and differentiation therapy. Mochizuki M, Nishimura R and Sasak N 2001. Carcinog. 21(7): 1271-1279. Effect of retinoids on growth inhibition of two canine melanoma cell lines. J Vet Med Sci. 63(1): Hong SH, Kadosawa T, Nozaki K, Mochizuki M, 83-86. Matsunaga S, Nishimura R and Sasaki N 2000. In Ohashi E, Miyajima N, Nakagawa T, Takahashi T, vitro retinoid-induced growth inhibition and Kagechika H, Mochizuki M, Nishimura R and morphologic differentiation of canine Sasaki N 2006. Retinoids induce growth osteosarcoma cells. Am J Vet Res. 61(1): 69-73. inhibition and apoptosis in mast cell tumor cell lines. J Vet Med Sci. 68(8): 797-802. Hong SH, Mochizuki M, Nishimura R, Sasaki N, Patnaik AK, MacEwen EG, Black AP and Luckow S Kadosawa T and Matsunaga S 2000. 1982. Extracutaneous mast-cell tumor in the dog. Differentiation induction of canine osteosarcoma Vet Pathol. 19(6): 608-615. cell lines by retinoids. Res Vet Sci. 68(1): 57-62. Patnaik AK, Ehler WJ and MacEwen EG 1984. Canine cutaneous mast cell tumor: morphologic grading Jark PC, Mundin DB, de Carvalho M, Ferioli RB, Anai and survival time in 83 dogs. Vet Pathol. 21(5): LA, Marchi FA, Rogatto SR, Laufer-Amorim R 469-474. and Tinucci-Costa M 2017. Genomic copy Pinello KC, Nagamine M, Silva TC, Matsuzaki P, number variation associated with clinical Caetano HV, Torres LN, Fukumasu H, Avanzo outcome in canine cutaneous mast cell tumors. JL, Matera JM and Dagli ML 2009. In vitro Res Vet SCi. 111: 26-30. chemosensitivity of canine mast cell tumors grades II and III to all-trans-retinoic acid (ATRA). Karayannopoulou M, Fytianou A, Assaloumidis N, Vet Res Commun. 33(6): 581-588. Psalla D and Savvas T 2013. Lipid peroxidation in Ricciarelli R, Zingg JM and Azzi A 2001. Vitamin E: neoplastic tissue of dogs with mammary cancer protective role of a Janus molecule. FASEB J. fed with different kinds of diet. Turk J Vet Anim 15(13): 2314-2325. Sci. 37: 449-453. Rizvi S, Raza ST, Ahmed F, Ahmad A, Abbas S and Mahdi F 2014. The role of vitamin e in human Katakwar P, Metgud R, Naik S and Mittal R 2016. health and some diseases. Sultan Qaboos Univ Oxidative stress marker in oral cancer: A review. Med J. 14(2): e157-165. J Can Res Ther. 12(2): 438-446. Saydam N, Kirb A, Demir O, Hazan E, Oto O, Saydam O and Guner G 1997. Determination of Kerksick C and Willoughby D 2005. The antioxidant glutathione, glutathione reductase, glutathione role of glutathione and N-acetyl-cysteine peroxidase and glutathione S-transferase levels in supplements and exercise-induced oxidative human lung cancer tissues. Cancer Lett. 119(1): stress. J Int Soc Sports Nutr. 2: 38-44. 13-19. Schweigert FJ and Bok V 2000. Vitamin A in blood Kumar A, Sharma S, Pundir CS and Sharma A 1995. plasma and urine of dogs is affected by the Decreased plasma glutathione in cancer of the dietary level of vitamin A. Int J Vitam Nutr Res. uterine cervix. Cancer Lett. 94(1): 107-111. 70(3): 84-91. Scott KC, Hill RC, Lewis DD, Boning AJ Jr and Lengfelder E, Reichert A, Schoch C, Haase D, Haferlach Sundstrom DA 2001. Effect of alpha-tocopheryl T, Loffler H, Staib P, Heyll A, Seifarth W, Saussele acetate supplementation on vitamin E S, Fonatsch C, Gassmann W, Ludwig WD, concentrations in Greyhounds before and after a Hochhaus A, Beelen D, Aul C, Sauerland MC, race. Am J Vet Res. 62(7): 1118-1120. Heinecke A, Hehlmann R, Wormann B, Sebbag L, Forrat R, Canet E, Renaud S, Delaye J and de Hiddemann W and Buchner T 2000. Double Lorgeril M 1994. Effects of dietary induction strategy including high dose supplementation with alpha-tocopherol on cytarabine in combination with all-trans retinoic myocardial infarct size and ventricular acid: effects in patients with newly diagnosed acute promyelocytic leukemia. German AML Cooperative Group. Leukemia. 14(8): 1362-1370. Lengfelder E, Saussele S, Weisser A, Buchner T and Hehlmann R 2005. Treatment concepts of acute promyelocytic leukemia. Crit Rev Oncol Hematol. 56(2): 261-274. Lewicka M, Henrykowska G, Zawadzka M, Rutkowski M, Pacholski K and Buczynski A 2017. Impact of electromagnetic radiation emitted by monitors on changes in the cellular membrane structure and protective antioxidant effect of vitamin A - In

6346 Macotpet A. et al. / Thai J Vet Med. 2018. 48(4): 631-637. arrhythmias in a dog model of ischemia- reperfusion. J Am Coll Cardiol. 24(6): 1580-1585. Shigesawa T, Sato C and Marumo F 1992. Significance of plasma glutathione determination in patients with alcoholic and non-alcoholic liver disease. J Gastroenterol Hepatol. 7(1): 7-11. Tang XH and Gudas LJ 2011. Retinoids, retinoic acid receptors, and cancer. Annu Rev Pathol. 6: 345- 364. Thurnham DI, Smith E and Flora P.S 1988. Concurrent liquid-chromatographic assay of retinol, alpha- tocopherol, beta-carotene, alpha-carotene, lycopene, and beta-cryptoxanthin in plasma, with tocopherol acetate as internal standard. Clin Chem. 34(2): 377-381. Traber MG and Atkinson J 2007. Vitamin E, antioxidant and nothing more. Free Radic Biol Med. 43(1): 4- 15. Valko M, Rhodes CJ, Moncol J, Izakovic M and Mazur M 2006. Free radicals, metals and antioxidants in oxidative stress-induced cancer. Chem Biol Interact. 160(1): 1-40. Waris G and Ahsan H 2006. Reactive oxygen species: role in the development of cancer and various chronic conditions. J Carcinog. 5:14. Welle MM, Bley CR, Howard J and Rufenacht S 2008. Canine mast cell tumours: a review of the pathogenesis, clinical features, pathology and treatment. Vet Dermatol. 19(6): 321-339. Winter JL, Barber LG, Freeman L, Griessmayr PC, Milbury PE and Blumberg JB 2009. Antioxidant status and biomarkers of oxidative stress in dogs with lymphoma. J Vet Intern Med. 23(2): 311-316.

Macotpet A. et al. / Thai J Vet Med. 2018. 48(4): 631-637. 663375 บทคัดยอ่ ภาวะเครียดออกซเิ ดชันและสารตา้ นอนมุ ลู อสิ ระในสุนขั ที่เป็นเนอื้ งอกมาสตเ์ ซลลท์ ี่ผิวหนงั อารยาพร มคธเพศ1* เอกชยั ภัทรพนั ธ์วเิ ชยี ร2 ฟ้านา่ น สขุ สวสั ดิ์1 พัชรี บุญศิริ3 ภาวะเครียดออกซเิ ดชนั เกดิ จากอนุมูลอิสระกลมุ่ ออกซิเจนมีปริมาณมากเกนิ ไป ระบบของสารต้านอนมุ ลู อิสระเกิดความเสียหาย หรือทง้ั สองภาวะรว่ มกัน ซึง่ ทาใหเ้ กดิ ความเสยี หายแก่สารชวี โมเลกุลประเภทไขมัน โปรตนี และดีเอน็ เอ การเปล่ยี นแปลงเหลา่ นีอ้ าจมีส่วน เกีย่ วข้องกับกระบวนการเกิดโรคมะเร็ง และมคี วามสมั พนั ธก์ บั มะเรง็ หลายชนิดในสุนขั วตั ถุประสงค์ของการศกึ ษาน้เี พ่ือเปรียบเทยี บระดบั ของมาลอนไดอัลดไี ฮด์ โปรตีนไฮโดรเปอร์รอกไซด์ กลูตาไธโอน เรตินอล และอลั ฟาโทโคฟีรอล ระหว่างสุนขั ทเ่ี ปน็ เนื้องอกมาสต์เซลล์ทผี่ ิวหนงั และสนุ ัขสขุ ภาพดี โดยการเกบ็ ตวั อย่างเลือดจากสนุ ัขสขุ ภาพดีจานวน 18 ตวั และสนุ ัขท่ีเป็นเนื้องอกมาสต์เซลลจ์ านวน 14 ตัว จากนนั้ นา ตัวอยา่ งเลือดไปตรวจหาระดับมาลอนไดอลั ดไี ฮด์ดว้ ยวิธีไทโอบาบิทูรคิ และระดับของโปรตนี ไฮโดรเปอรร์ อกไซดด์ ว้ ยวิธเี ฟอริค-ไซลนิ อล ออ เรนจ์ การวดั ระดับกลูตาไธโอนใช้วธิ กี ารวัดค่าการดูดกลนื แสง สว่ นการตรวจวดั ระดับเรตินอลและอัลฟาโทโคฟรี อลใช้วิธโี ครมาโทกราฟี ของเหลวสมรรถนะสูง พบวา่ สนุ ัขที่เปน็ เนื้องอกมาสต์เซลลม์ รี ะดับของมาลอนไดอลั ดีไฮดแ์ ละระดบั ของโปรตีนไฮโดรเปอร์รอกไซดส์ งู กว่าสุนัข สุขภาพดอี ยา่ งมนี ัยสาคญั (P<0.01 และ P<0.05 ตามลาดบั ) แตม่ ีระดับของกลตู าไธโอน ระดบั ของเรตนิ อล และระดับของอัลฟาโทโคฟีรอลมี ค่าต่ากวา่ สุนขั สขุ ภาพดอี ย่างมีนัยสาคญั (P<0.01, P<0.05 และP<0.01 ตามลาดับ) ดงั นน้ั สนุ ัขที่ป่วยเปน็ เนอ้ื งอกมาสต์เซลลอ์ าจมีการพฒั นา หรือเกย่ี วข้องกบั ภาวะเครยี ดออกซเิ ดชัน เนื่องมาจากมีภาวะความไมส่ มดลุ กันระหว่างภาวะเครียดออกซิเดชนั และความสามารถของสารตา้ น ออกซเิ ดชันในร่างกาย ทงั้ นคี้ วรมกี ารศกึ ษาเพิม่ เติมในเรื่องของภาวะเครียดออกซิเดชันและฤทธิ์ของสารตา้ นอนมุ ลู อิสระในสุนัข เพ่ือเป็น แนวทางทอ่ี าจเปน็ ไปได้ในการวางแผนสาหรบั เปน็ การรกั ษาเสรมิ ในสนุ ัขทเี่ ป็นมะเร็งตอ่ ไปในอนาคต คาสาคัญ: กลูตาไธโอน โปรตีนไฮโดรเปอรร์ อกไซด์ เนือ้ งอกมาสตเ์ ซลล์ มาลอนไดอลั ดไี ฮด์ เรตนิ อล สุนัข อัลฟาโทโคฟรี อล 1กลมุ่ วิชาอายรุ ศาสตรส์ ัตวเ์ ลย้ี ง คณะสตั วแพทยศาสตร์ มหาวทิ ยาลยั ขอนแกน่ จงั หวดั ขอนแกน่ ประเทศไทย 40002 2กล่มุ วิชาพยาธิชวี วทิ ยา คณะสัตวแพทยศาสตร์ มหาวิทยาลยั ขอนแกน่ จงั หวัดขอนแกน่ ประเทศไทย 40002 3ภาควิชาชีวเคมี คณะแพทยศาสตร์ มหาวิทยาลัยขอนแก่น จงั หวดั ขอนแกน่ ประเทศไทย 40002 *ผู้รบั ผิดชอบบทความ E-mail: [email protected]



Original Article Impact of feed characteristics for monitoring AFB1 contamination in commercial concentrate feeds used in small holder dairy farms, Chiangmai, Thailand Wasana Chaisri1 Wantanwa Mongkon1 Yoshigo Sugita-Konishi2 Witaya Suriyasathaporn1* Abstract The objective of this study was to determine the impact of abnormal physical characteristics for monitoring aflatoxin B1 (AFB1) contamination in commercial dairy concentrate feeds. The study was conducted during March to May 2013 using 82 dairy farms in Chiang Mai province, Thailand. All feed samples were collected from new packages and their physical characteristics were subsequently determined. AFB1 concentrations in the samples were measured using the enzyme-linked immunosorbent assay (ELISA), and feed samples with AFB1 of more than 5 µg/Kg were determined as AFB1 contaminated samples. The impact of physical characteristics on AFB1 contamination were evaluated using a multiple logistic model with a p-value less than 0.05. Forty percent (39 of 97) of the feed samples were contaminated with AFB1. Concentrate feeds with rancid or sour odors had a higher risk of AFB1 contamination (OR =4.79, P=0.03), as did samples with a high appearance of a brightly illuminated fluorescence on feed after testing with black light (OR= 24.04, P = 0.0002). Farmers can use these indicators as screening tests before purchasing the feed. Keywords: aflatoxin B1; concentrates; contamination; dairy cow; physical characteristics, black light 1Department of Food Animal Clinics, Faculty of Veterinary Medicine, Chiang Mai University, Mueang, Chiang Mai, 50100, Thailand 2Department of Food and Life Sciences, The Graduate School of Life and Environmental Sciences, Azabu University, 1-17-71, Fuchinobe Chuo-ku, Sagamihara, Tokyo, 2298501, Japan *Correspondence: [email protected] Thai J Vet Med. 2018. 48(4): 639-644.

640 Chaisri W. et al. / Thai J Vet Med. 2018. 48(4): 639-644. Introduction Physical feed characteristics The physical characteristics of feed samples were established by the Aflatoxin M1 (AFM1), a carcinogenic same investigator, including proportions of cracked mycotoxin in dairy products, is a serious concern as the particles, presence of abnormal odor, presence of consumption of these products poses health risks, abnormal color, proportion of pack particle and particularly to young people (Pradini et al., 2009). In proportion of bright fluorescence after testing with Thailand, many reports have shown high levels of black light. Based on a normal pellet length of 2 cm, AFM1 in both raw milk and drinking milk cracked particles were defined as a pellet length ≤1 cm (Suriyasathaporn & Nakprasert, 2012; Chaisri et al., by visual observation. Samples were categorized into 2017). The main reason for AFM1 contamination is the three levels: not cracked (less than 10% cracked consumption of feed contaminated with AFB1 particles); partially cracked (10 to 50% cracked (Bantaokul & Ruangwises, 2010). Not only causing particles); and most cracked (more than 50% cracked AFM1 contamination in milk, AFB1 is also the most particles). Sour or rancid odors were defined as toxic mycotoxin metabolite causing a reduction of feed abnormal odor of the feeds. The normal color of dairy consumption, growth retardation, reduced concentrate feed is yellow or light brown, dependent performance, immune suppression, irritated tissues, on the raw materials. Darker than normal, dark brown causing abortions and death (Hall and Wild, 1994). or black were considered abnormal colors. The pack Because of the potential hazards of this toxin, many particle was the packed characteristic of the feed countries have set guidelines of acceptable levels in sample after hand squeezing. The procedure for animal feed to reduce the risk of AFM1 contamination fluorescence light testing for feed samples has been in milk. The regulation in Europe and the United states described by Mongkon et al. (2014); a positive result limits AFB1 in animal feed to 5 and 20 µg/Kg was defined as a brightly illuminated fluorescence (European Food Safety Authority, 2004), while the appearing on the pellet area. Samples were categorized regulation in Thailand limits total aflatoxin to 100 as negative, partly positive and highly positive to black µg/Kg and 200 µg/Kg for young and mature animals, light tests when the percentages of positive pellets respectively (Thai Ministry of Agriculture and were 0%, 0-10%, or ≥10% of all pellets, respectively. Cooperative regulation, 2016). Aflatoxin B1 analysis Each feed sample was In Thailand, most farmers feed their cows with a classical mixed ration of roughage and carefully mixed and finely ground, and then extracted concentrates (Mongkon et al., 2014; Oberheu & Dabbert, 2011). For convenience, many dairy farmers by putting 20 g of the ground sample into an purchase commercial concentrate feeds (Mongkon et al., 2014; Oberheu & Dabbert, 2011; Mongkon et al., Erlenmeyer flask to which 100 ml of 70% methyl 2017; Suriyasathaporn & Nakprasert, 2012). Regardless of farm management factors, especially storage alcohol was added. The flask was shaken at 300 rounds practices, related to AFM1 contamination (Mongkon et al., 2014; Suriyasathaporn & Nakprasert, 2012), the per min for 30 mins, after which the mixture was easiest way of preventing AFM1 contamination in milk is to purchase feed with minimal AFB1 contamination. filtered by Whatman no. 4 paper. The AFB1 level was Abnormal characteristics of commercial concentrate feeds might be used to monitor AFB1 contamination. measured by DOA-Aflatoxin ELISA test kit Therefore, the objective of this study was to determine the impact of feed characteristics for monitoring AFB1 (Postharvest and Processing Research and contamination in commercial dairy concentrate feeds. Development Division, Thailand). The kit had a Materials and Methods recovery between 82-100 % and a sensitivity limit of 0.4 Chemicals and Reagents All reagents were of analytical grade. Methyl alcohol was purchased from µg/Kg (Chinaphuti et al., 2002). The analysis was Merck (Darmstadt, Germany). DOA-Aflatoxin ELISA test kit (Department of Agriculture, Ministry of performed according to the kit instruction. Briefly, 50 Agriculture Cooperatives, Thailand) was purchased from Higher Enterprises Co.Ltd, Pathum Thani, μl of either AFB1 standard (0, 0.2, 0.5, 1, 2 ng/ml) or the Thailand. diluted samples were added into antibody coated Study design and feed collection A cross-sectional study was performed during March to May 2013 using wells; then 50 μl of Aflatoxin B1-Horseradish all 82 smallholder dairy farms from a dairy cooperative in Chiang Mai province, Thailand. To minimize factors Peroxidase (AFB1-HRP) conjugate was added to each relating to farm management on AFB1 contamination of commercial dairy concentrate feed, all samples were well, slightly shaken and incubated at room collected from new packages of all concentrate feeds using in farms. One kilogram of each sample was temperature for 30 mins. The contents of the well were collected, kept on ice in a plastic zip lock bag, transported to the laboratory and stored in a freezer dumped into the appropriate waste container and the (-18oC) until analysis. plate was washed 3 to 5 times by 0.01M phosphate buffer saline with 0.5% Tween 20 (PBS-T). Tetramethylbenzidine substrate (100 μl) was added to each well and incubated for 10 mins at room temperature. The reaction was stopped by adding 100 μl of 0.3M phosphoric acid. The solutions were read at 450 nm using an automated MicroELISA spectrophotometer reader. The results of absorbance were expressed as percentages of maximal binding as follows: % maximal binding = B/B0 x 100 Where B= mean absorbance of feed sample and B0= mean absorbance of AFB1 standard at 0 ng/ml The standard curve was generated by plotting the concentration of AFB1 with % maximal binding in Log base 10 values as shown in Fig 1. To test the recovery of the DOA-Aflatoxin ELISA method based on the

Chaisri W. et al. / Thai J Vet Med. 2018. 48(4): 639-644. 641 standard curve, concentrate feed samples were spiked AFB1. For univariable analysis, the impact of feed with AFB1 at a concentration of 50 µg/Kg, selected as characteristics on AFB1 contamination were separately the half concentration of a previous recovery test study determined using Fisher’s exact chi-square test. The (Chinaphuti et al., 2002). final model of impacts of feed characteristics on AFB1 contamination was determined using a multiple Statistical analysis Data was described as frequencies logistic model with a backward selection method (SAS for categorical variables. Feed samples with AFB1 at Institute, 1997). All factors were first entered and concentrations higher than 5 µg/kg, the regulatory subsequently remained in the model when their P- limit for dairy animal feed of the European Food Safety values, indicated by the likelihood ratio tests, were less Authority (EFSA) (European Food Safety Authority, than 0.10. 2002), were determined as samples contaminated with Log% maximum binding 2.0 R² = 0.9834 1.8 1.6 0.5 1 1.5 2 1.4 Concentration (ng/ml) 1.2 1.0 0.8 0.6 0.4 0.2 0.0 0 Figure 1 The enzyme-linked immunosorbant assay (ELISA) standard curve for the determination of AFB1 concentration Results From the final analysis, the factors having impact on AFB1 contamination included abnormal The accuracy of AFB1 measurement was odor and having a brightly illuminated light under measured by the ELISA standard curve (Fig. 1). The black light test (Table 2). The concentrates with a sour correlation coefficient (R²) at 0.98 indicated a good or rancid odor had a higher risk of AFB1 contamination validity of the AFB1 measurement. The test was than those with a normal odor (OR = 4.79, P = 0.03). repeated five times and the recovery rate was 88.4%. Feed samples with more than10% of brightly lit area or highly positive had a higher risk of AFB1 The mean ± SD of AFB1 contamination of all contamination than samples without any bright area 97 feed samples from 82 farms, in which 15 farms used (OR = 24.04, P = 0.0002). 2 concentrate feeds and 67 farms had only 1 concentrate feed, was 5.41 ± 3.61 µg/kg, ranging from Discussion 1.10 to 23.61 µg/Kg. Forty percent of the samples (n = 39) exceeded the EFSA allowable limit, and were In the last decade, many countries in considered contaminated with AFB1. The tropical regions have encountered aflatoxin characteristics of the commercial concentrates are contamination problems. In the present study, we shown in Table 1. The commercial concentrates found approximately 40% of feed samples were exhibited high levels of abnormal or improper contaminated with AFB1. The high incidence of AFB1 characteristics including abnormal color 12.4% (12/97), contamination found in this study was in line with sour or rancid odor 19.6% (19/97), abnormal pack previous studies in other tropical countries (Arunvipas particle 27.8% (27/97), abnormal cracked particles et al., 2009; Meemark & Sakdinun, 2006). Meemark & 43.3% (42/97) and positive to black light test 44.3% Sakdinun (2006) found that aflatoxin B1 was detected (43/97). From univariable analysis, 3 out of 5 tested in 287 feed samples (78.42%) ranging from 0.40-23.97 characteristics were significantly associated with AFB1 µg/Kg. Likewise, in this study, AFB1 concentrations contamination (P<0.01) including cracked particles, found in concentrated feed ranged between 1.105 to abnormal odor and positive results having bright 23.61 µg/Kg; these results were consistent with ranges illumination under black light test. The highest previously reported in Thailand (Charoenpornsook & percentages of AFB1 contamination (44.3%) were Kavisarasai, 2014). Due to the uses of ELISA, instead of found in samples with the most cracked particles the uses of high performance liquid chromatography (93.8%), highly positive under black light test (87%) as a standard method for aflatoxin measurement, and sour or rancid odors (73.7%). aflatoxin contamination parameter, defined at AFB1>5

642 Chaisri W. et al. / Thai J Vet Med. 2018. 48(4): 639-644. µg/Kg, were used instead of their AFB1 humidity and fungal growth (Martinez & Woloshuk, concentrations. 2012; Moreira et al., 1996). Among the abnormal characteristics examined in this study, cracked or In this study, abnormal commercial damaged feed pellets were the only factors caused by concentrate characteristics were found in high a physical mechanism (Dorner et al., 1989; Gizachew et percentages, ranging between 12.4% (12/97) to 44.3% al., 2016). The other characteristics we examined (43/97), as shown in Table 1. In general, dairy required biological processes. Fungi and mycotoxin concentrate feed pellet characteristics, including length contamination, including AFB1, could occur in the raw or particle size, depend on the manufacturing process materials before manufacturing (Das et al., 1993; (Bryden, 2012; Gizachew et al., 2016). After production, Gizachew et al., 2016). feed pellets are stored in plastic bags to prevent Table 1 Association of feed characteristics of concentrate feed samples (n = 97) with AFB1 contamination (indicated by an AFB1 concentration above 5 µg/kg, the allowable contamination limit in complete or concentrate feed for dairy cattle as defined by EFSA, 2002) Feed characteristic Total AFB1 contaminated sample (n = 39) P-value (n = 97) <0.001 Cracked particle Number Percentage (%) Mostly cracked 16 0.0014 Partially crack 26 15 93.8 0.2142 Not cracked 55 9 34.7 0.3609 15 27.3 <0.001 Odor 19 Sour or rancid 78 14 73.7 Normal 25 32.1 12 Color 85 7 58.3 Abnormal 32 37.7 Normal 27 70 13 48.2 Pack particle 26 37.1 Abnormal 23 Normal 20 20 87 54 8 40 Black light test 11 20.4 Mostly positive Partially positive Negative Table 2 The final model indicating physical feed characteristics associated with AFB1 contamination in commercial dairy concentrate feed Feed characteristic Estimate SEa P-value Odds ratio 95% CIb Odor 1.57 0.73 0.03 4.79 1.15 – 19.94 Sour or rancid --------------------------------Reference----------------------------- Normal 1.15 0.66 0.08 3.17 0.85 – 11.77 Black light test Partially positive 3.18 0.85 0.0002 24.04 4.54 – 127.18 Mostly positive Negative --------------------------------Reference----------------------------- aSE, Standard error b95% CI, 95% Confidence interval From simple analyses, cracked particles, a corn kernels (Abbas et al., 2004; Adams et al., 1993). sour or rancid odor, and a positive black light test Grain or feed spoiled by fungal growth had an significantly increased the likelihood of AFB1 abnormal odor, including musty, sour or putrid smell contamination (Table 1). However, the cracked particle (Atanda et al., 2016; Miller, 1995). Humidity in cracked variable was not included in the final model of this feed pellet particles, either during manufacturing or study (Table 2), in contrast to our previous study that later storage, might promote fungal growth and AFB1 included cracked particles (as well as a positive black production (Atanda et al., 2016; Das et al., 1993). In light test) as a risk factor for AFM1 contamination in addition, cracked pellets are more likely to be milk (Mongkon et al., 2017). The luminescence or contaminated with fungi (Adams et al., 1993). positive black light result indicated the presence of fungi, especially AFB1-producing fungi (Moreira et al., In conclusion, the physical feed 1996), and was related to aflatoxin contamination of characteristics especially abnormal odors and a positive black light test, were associated with AFB1

Chaisri W. et al. / Thai J Vet Med. 2018. 48(4): 639-644. 643 contamination in dairy concentrate feeds, and can be as undesirable substance in animal feed. ESFA. used as factors for monitoring AFB1 contaminated feeds especially before purchase by farmers. 2(3): 1-27. Acknowledgements Gizachew D, Szonyi B, Tegegne A, Hanson J and Grace The authors thank the dairy farmers of the D 2016. Aflatoxin contamination of milk and Patung Dairy Cooperative, Chiang Mai for their participation and appreciate the Faculty of Veterinary dairy feeds in the Greater Addis Ababa milk Medicine, Chiang Mai University for supporting. shed, Ethiopia. Food Control. 59: 773-779. References Hall AJ, Wild CP 1994. Epidemiology of aflatoxin Abbas HK, Shier WT, Horn WT and Weaver MA 2004. Cultural Methods for Aflatoxin Detection. J related disease. In: The Toxicology of Aflatoxins: Toxicol Toxin Rev. 23(2-3): 295-315. Human Health, Veterinary and Agricultural Adams RS, Kephart KB, Ishler VA, Hutchinson LJ and Roth GW 1993. “Mould and mycotoxin problems Significance 1st ed D Eaton and J Groopman (ed) in livestock feeding.” Dept of Dairy and Animal Science, Extension Publ. DAS 93–21. [Online]. San Diego CA: Academic Press 233–258. Available:1-17. http://crbh.psu.edu/das/research- Martinez EM, & Woloshuk C 2012. Monitoring for extension/dairy/nutrition/pdf/mold.pdf. Spoilage and Mycotoxins. In Stored Product Arunvipas P, Prapeuk T and Jala P 2009. Study on risk factors of aflatoxin B1 contamination in dairy feed Protection 1st ed(ed) Kansus: K-State Research in Kanchanaburi, Nakhon Pathom and Ratchaburi provinces. J Thai Vet Med Assoc. 60(1- and Extension 283-288 3): 39-48. Memark N and Sukdinan P 2006. Aflatoxin in Dairy Atanda SA, Pessu PO, Agoda S, Isong IU, Adekalu OA, Echendu MA and Falade TC 2016. Fungi and Feed from Western Part of Thailand. Thai-NIAH mycotoxins in stored foods. Afr J Microbiol Res. 5(25): 4313-4382. eJournal ISSN 1905-5048, 162-170. Available: Bryden WL 2012. Food and feed, mycotoxins and the http://www.dld.go.th/niah. Accessed perpetual pentagram in a changing animal production environment. Anim Prod Sci. 52(6-7): November 20, 2017 383-397. Miller JD 1995. Fungi and Mycotoxins in Grain: Bantaokul C, Ruangwises S 2010. Carry-over rate of aflatoxin M1 into cow milk during early lactation Implications for Stored Product Research. J period. Proc 9th CU Vet Sci Ann Con. 128 Stored Prod Res. 31(1): 1-16. Chaisri W, Mongkon W, Sugita-Konishi Y, Van Dam D, Huntley I, Suriyasathaporn W 2017. Feed and The Ministry of Agriculture and Cooperative 2016. feed storage factors in relation to aflatoxin M1 contamiantion in bulk milk of smallholder dairy Thai ministry of agriculture and cooperative farms. JSM Mycotoxins. 67(2): 1-4. regulation. [Online]. Available: Charoenpornsook K and Kavisarasai P 2014. Determination of aflatoxin B1 in food products in http://afvc.dld.go.th/index.php. Accessed July Thailand. Afr J Biotechnol. 13(53): 4761-4765. 2018. Chinaphuti A, Trikarunasawat C, Wongurai A and Kositcharoenkul S 2002. Production of In-house Mongkon W, Poapolathep A, Kumagai S and ELISA Test Kit for Detection of Aflatoxin in Agricultural Commodities and Their Validations. Suriyasathaporn W 2014. Use of blue-greenish Kasetsart J (Nat Sci). 36:179 – 186. yellow fluorescence test on feeds and its Das HK, Hattula MT, Myllymi OM and Miilkki Y 1993. Effects of formulation and processing variables association with aflatoxin m1 contamination in on dry fish feed pellets containing fish waste. J Sci Food Agric. 61(2): 181-191. bulk tank milk. J Food Pro. 77(2): 284-291. Dorner JW, Cole RJ, Sanders TH and Blankenship PD Mongkon W, Sugita-Konishi Y, Chaisri W and 1989. Interrelationship of kernel water activity, soil temperature, maturity, and phytoalexin Suriyasathaporn W 2017. Aflatoxin B1 production in preharvest aflatoxin contamination of drought-stressed peanuts. Mycopathologia. contamination of dairy feeds after storage in farm 105(2): 117-128. practice in tropical environment. Biocontrol Sci. European Food Safety Authority 2004. Opinion of the scientific panel on contaminants in the food chain 22(1): 41-45. at a request from the commission related to AFB1 Moreira MT, Sanroman A, Feijoo G and Kema JM 1996. Control of pellet morphology of filamentous fungi in fluidized bed bioreactors by means of a pulsing flow. Application to Aspergillus niger and Phanerochaete chrysosporium. Enzyme Microb Technol. 19(4): 261-266 Oberheu DG and Dabbert CB 2011. Aflatoxin production in supplemental feeders provided for northern bobwhite in Texas and Oklahoma. J Wildl Dis. 37(3), 475-480. 37(3): 475-480. Prandini A, Tansini G, Sigolo S, Filippi I, Laporta M and Piva G 2009. On the occurrence of aflatoxin Ml in milk and dairy products. Food Chem Toxicol. 47(5): 984-991. SAS Institute, Inc. SAS/STAT software: changes and enhancements through release 6.12. Cary, Carolina, United States, 1997 Suriyasathaporn W and Nakprasert W 2012. Seasonal patterns of aflatoxin M1 contamination in commercial pasteurize milk from different areas in Thailand. Food Addit Contam Surveill Part B. 5(2): 145-149.

644 Chaisri W. et al. / Thai J Vet Med. 2018. 48(4): 639-644. บทคดั ยอ่ ผลของลักษณะอาหารเพือ่ เฝ้าระวงั การปนเปอ้ื นอะฟลาทอกซนี บี 1 ในอาหารขน้ ทีม่ จี าหนา่ ยทใี่ ชใ้ นฟารม์ โคนมรายย่อยในเชยี งใหม่ ประเทศไทย วาสนา ไชยศรี1 วนั ธนั วา มงคล1 โยชิโกะ ซกู ิตะ-โคนชิ ิ2 วทิ ยา สุริยาสถาพร1* วตั ถปุ ระสงค์ของการศึกษาน้เี พื่อหาผลกระทบของลกั ษณะทางกายภาพทผ่ี ิดปกติเพอ่ื ช่วยในการตรวจสอบการปนเปื้อนอะฟลา ทอกซนิ บี 1 (AFB1) ในอาหารสาเร็จรปู โคนม ดาเนินการศกึ ษาระหวา่ งเดอื นมีนาคมถงึ พฤษภาคม พ.ศ. 2557 โดยใช้ฟารม์ โคนมในจงั หวดั เชยี งใหมจ่ านวน 82 ฟารม์ เกบ็ ตวั อยา่ งอาหารและสงั เกตลักษณะทางกายภาพของอาหารหลงั จากเปิดถงุ อาหารทนั ที วดั ปรมิ าณอะฟลาทอก ซนิ บี 1 โดยใช้วธิ ี Enzyme-linked immunosorbent assay (ELISA) และตวั อยา่ งอาหารที่มอี ะฟลาทอ็ กซินเกิน 5 ไมโครกรัม/กโิ ลกรับถูก กาหนดเป็นตัวอยา่ งท่มี กี ารปนเป้ือน AFB1 ทดสอบความสัมพนั ธร์ ะหว่างลักษณะทางกายภาพของอาหารและการปนเปอื้ นอะฟลาทอกซนิ โดย ใช้ multiple logistic model และกาหนดคา่ p-value เท่ากบั 0.05 พบวา่ ตวั อย่างอาหารรอ้ ยละ 40 (39 จาก 97 ตัวอยา่ ง) ปนเป้ือนอะฟ ลาทอกซินบี 1 อาหารข้นที่มกี ลิ่นเหม็นหืนหรอื เหมน็ เปรย้ี วเส่ยี งตอ่ การปนเปื้อนอะฟลาทอกซนิ มากกว่าอาหารทมี่ กี ลิน่ ปกติ (Odds ratio (OR)=4.79, P=0.03) เช่นเดยี วกับตวั อย่างอาหารท่เี รืองแสงฟลูออเรสเซนต์หลังจากทดสอบดว้ ยหลอด black light (OR=24.04, P=0.002) เกษตรกรผเู้ ล้ยี งโคนมสามารถใช้ลกั ษณะดังกล่าวเพือ่ คัดกรองอาหารก่อนซอ้ื เขา้ ฟารม์ ได้ คาสาคญั : อะฟลาทอกซินบี 1 อาหารข้น การปนเปือ้ น โคนม ลกั ษณะทางกายภาพ แบลค็ ไลต์ 1ภาควชิ าคลินกิ สัตว์บรโิ ภค คณะสตั วแพทยศาสตร์ มหาวิทยาลยั เชียงใหม่ อ.เมือง จ.เชียงใหม่ ประเทศไทย 50100 2ภาควชิ าอาหารและวทิ ยาศาสตรช์ วี ภาพ บัณฑิตวิทยาลัยวทิ ยาศาสตรช์ ีวภาพและสิง่ แวดลอ้ ม มหาวทิ ยาลยั อาซาบุ 1-17-71 ฟชู ิโนเบะ ชโู อ-กุ ซากามิฮาระ โตเกยี ว ประเทศญ่ีปุ่น 2298501 *ผูร้ บั ผิดชอบบทความ E-mail: [email protected]

Original Article Comparative effect of DL-methionine and DL-methionine hydroxy analogue supplemented diet on productive performance, fat accumulation and lipid profile in blood of meat-type ducks Chanwit Kaewtapee1* Chaiyapoom Bunchasak1 Abstract The objective of this study was to compare the effect of DL-methionine (DLM) and DL-methionine hydroxy analogue (DL-MHA) supplementation on the growth performance, carcass quality, fat accumulation and lipid profile in blood of meat-type duck. Three hundred male Cherry Valley ducks were divided into 3 groups with 4 replications of 25 ducklings each. Each group was given a diet as follows: 1) basal diet, 2) basal diet + DLM, and 3) basal diet + DL- MHA. The results show that dietary supplementation with DLM and DL-MHA increased body weight (P<0.01), feed intake (P<0.01) and pectoralis major (P<0.05), but decreased abdominal fat (P<0.01) when compared to the basal diet. Supplementation with DL-MHA also resulted in greater levels of triglycerides (P<0.05), low-density lipoprotein (P<0.05) and cholesterol (P<0.05) in the blood compared to the DLM group, in addition to greater levels of triglycerides (P<0.05) and high-density lipoprotein (P<0.05) compared to the basal diet. Furthermore, supplementation with DLM was shown to lower the level of plasma uric acid (P<0.05) compared to the basal diet. It is concluded that DL-MHA can promote the same productive performance in meat-type ducks as DLM, but the fat accumulation and lipid profile in blood was greater with the DL-MHA supplementation than with DLM. Furthermore, the underlying mechanism of different effects between DL-MHA and DLM on the lipid metabolism in ducks should be further investigated. Keywords: DL-methionine, DL-methionine hydroxy analogue, fat accumulation, meat-type duck, productive performance 1Department of Animal Science, Faculty of Agriculture, Kasetsart University, 50 Ngam Wong Wan Rd, Lat Yao, Chatuchak, Bangkok 10900 *Correspondence: [email protected] Thai J Vet Med. 2018. 48(4): 645-653.

646 Kaewtapee C. and Bunchasak C. / Thai J Vet Med. 2018. 48(4): 645-653. Introduction procedures were conducted according to the regulations of the Institute of Animal for Scientific In poultry nutrition, methionine (Met) is Purposes Development. Three hundred male, meat- considered as the first-limiting amino acid, particularly type ducks (Cherry Valley) were used in the in a corn-soybean based diet (Jianlin et al., 2004). Due experiment. The ducklings were divided into 3 groups to the sparing effect of cystine, Met can be with 4 replications of 25 ducklings each. Floor pens (0.2 supplemented in diets to meet the total sulfur amino m2 per bird) were located in an evaporative cooling acid (SAA) requirement for poultry (Bunchasak, 2009). system. The temperature was set at 33C at one-day- DL-methionine (DLM) and DL-methionine hydroxy old and then was decreased by 1C at 3-day intervals analogue (DL-MHA) are forms of synthetic Met, which until a final temperature of 25C was reached. The are widely used in livestock feed (Bunchasak, 2014). lighting and vaccination were managed according to The supplementation of Met in the diet of white Peking the recommended requirements for the strain. Each ducks has an enormous impact on growth performance pen was equipped with two hanging feeders and six and carcass quality (Xie et al., 2006), while the Met nipple drinkers. Feed and water were provided ad requirement in modern white Peking ducks seems to libitum throughout the period of the study. be higher than that of the NRC (1994) recommendation (Xie et al., 2004). Furthermore, the biological efficacy of Experimental design and diets: A completely DL-MHA in meat-type ducks for weight gain was randomized design was used. Three experimental similar to that of DLM when it was included in the diet diets (pellet form) were provided as shown in Table 1 as the same Met equivalent (Kluge et al., 2016). With as follows: 1) basal diet, 2) supplementation with regard to commercial feed ingredients or maximal powder form of DLM (Sumitomo Chemical Co., Ltd, SAA requirement, it has been recently suggested that Tokyo, Japan) to meet the recommended requirements the bioefficacy of DL-MHA was 88% (wt/wt) of the for Cherry Valley ducks (Cherry Valley Farms, 2004) at value for DLM, on a product-to-product basis, for levels of 2.6, 2.6, 3.2 and 3.6 g/kg for 0 to 9, 10 to 16, 17 growth performance of broiler chickens (Bunchasak to 42 and 43 to 47 days of age, respectively, and 3) and Keawarun, 2006). However, the Met or SAA supplementation with a liquid form of DL-MHA (88% requirements in meat-type ducks are generally lower of bioefficacy of the DLM on a product-to-product than those requirements in broiler chickens (Cherry basis, wt/wt; Sumitomo Chemical Co., Ltd, Tokyo, Valley Farm, 2004; Ross 308, 2014; Grimuad Freres, Japan) to meet the recommended requirements for 2010). Cherry Valley ducks (Cherry Valley Farms, 2004) at levels of 3.0, 3.0, 3.6 and 4.1 g/kg for 0 to 9, 10 to 16, 17 The function of methionine is closely related to 42 and 43 to 47 days of age, respectively. to lipid metabolism (Wong et al., 1977; Tillman and Pesti, 1986; Chen et al., 1993). In broiler chickens, liver Chemical analyses: Official standard methods (AOAC, triglycerides and fat accumulation are enhanced in 2000) were used to determine the contents of crude response to SAA supplementation (Bunchasak and protein. The crude protein was calculated as N × 6.25. Silapasorn, 2005), as Met may disturb the ability of Concentrations of lysine, methionine, cystine and lipid transportation (Bunchasak et al., 1997). In terms threonine in the basal diet were determined using ion- of Met source, the effect of DLM on fat metabolism in exchange chromatography with post-column broiler chickens is different from that of DL-MHA derivatization with ninhydrin (Llames and Fontaine, (Esteve-Garcia and Llauradó, 1997, Bunchasak and 1994). Tryptophan was determined using HPLC with Keawarun, 2006) and efficiency in the anti-lipogenic fluorescence detection (extinction 280 nm, emission 356 role of Met has been reported to be lower in DL-MHA nm), after alkaline hydrolysis with barium hydroxide than in DLM (Esteve-Garcia and Llauradó, 1997). octahydrate for 20 h at 110°C according to the Higher subcutaneous fat accumulation in ducks has procedure outlined by Commission Directive (2000). been shown to provide insulation and tolerance to low Total sulfur amino acids were calculated using temperatures (Bochno et al., 2013), whereas their methionine plus cystine. The calculated metabolizable abdominal fat accumulation is lower than in broiler energy and analyzed chemical composition are shown chickens (Sakulthai, 2013). The differences of some in Table 1. metabolic pathways involving fat metabolism between these species have been investigated (Mooney and Growth performance and carcass quality: All ducks Lane, 1981; Hermier, 1997; Jin et al., 2001). It is were weighed individually at the start and end of each hypothesized that due to the unique nature of fat phase. Feed intake was also recorded at the end of each accumulation and physiological functions of ducks, the phase. The average daily gain (ADG) and feed bioefficacy of DL-MHA compare to DLM may be conversion ratio (FCR) were calculated on a per pen different from other animal species. Therefore, the basis. The protein intake and SAA intake were objective of this research was to study the effects of calculated from the feed intake. At the end of the DLM in comparison to DL-MHA on the growth experiment (47 days of age), after overnight feed performance, carcass quality and lipid profile in blood deprivation, all the ducks were weighed. One duck of meat-type duck. from each pen was sacrificed by asphyxiation using CO2 in an atmosphere of less than 2% oxygen (air Materials and Methods displaced by CO2) for 1.5-2.0 mins. After the carcasses had been chilled at 4oC for 4 h, skin with subcutaneous Animals and management: The research was fat covering the pectoralis major area was removed. performed at the Animal Research Station The pectoralis major (outer breast meat), pectoralis (Department of Animal Science, Faculty of Agriculture, Kasetsart University, Bangkok), and all animal care

Kaewtapee C. and Bunchasak C. / Thai J Vet Med. 2018. 48(4): 645-653. 647 minor (inner breast meat), drumette, leg meat pad, including the fat surrounding the gizzard, were (including thigh and drum stick) and abdominal fat removed manually and weighed. Table 1 Basal diet (g/kg, as fed-basis) and chemical composition in each period of growth. Ingredient Starter 1 Starter 2 Grower Finisher (0 to 9 d) (10 to 16 d) (17 to 42 d) (43 to 47 d) Corn (8.1% CP) 299.3 299.3 298.1 297.8 Corn starch 3.0 3.0 3.6 4.1 Broken rice (7.7% CP) 35.5 129.7 106.7 144.0 Rice bran (13.8% CP) 129.2 27.1 90.2 100.0 Defatted rice bran (14.4% CP) - - 9.8 - Wheat bran (14.8% CP) 145.6 200.0 200.0 200.0 Oil Soybean meal (47.2% CP) 40.0 40.0 40.0 39.4 304.4 256.7 213.3 176.5 Limestone 16.7 16.3 19.8 20.4 Mono-dicalcium phosphate 15.8 17.3 9.2 8.1 (21.0% P) Sodium chloride 4.2 4.2 4.2 4.2 Choline chloride (75.0%) 0.8 0.8 0.8 0.8 Antimold 0.5 0.5 0.5 0.5 Antioxidant 0.1 0.1 0.1 0.1 Vitamin and trace mineral 1.5 1.5 1.5 1.5 premix1 DL-Methionine -- - - L-Lysine-HCL 2.6 2.3 1.4 1.2 L-Threonine 0.8 1.2 0.8 1.4 Chemical composition (g/kg, as dry matter basis) Metabolizable energy2 (kcal/kg) 2,850 2,900 2,900 2,950 Crude protein 220.0 200.0 185.0 169.9 Methionine 2.9 2.7 1.3 0.9 Total sulfur amino acids3 6.4 5.8 4.3 3.4 Lysine 13.5 11.7 10.0 8.8 Threonine 9.0 8.5 7.5 7.5 Tryptophan 2.6 2.4 2.2 2.0 1Vitaimin and mineral premix content (per kg of diet): retinyl acetate 4.13 mg, cholecalciferol 75 μg, α-tocopherol acetate 13.5 mg, vitamin K3 1.5 mg, vitamin B1 1.5 mg, vitamin B2 5 mg, vitamin B6 2 mg, vitamin B12 0.05 mg, niacin 25 mg, Ca-D-panthothenate 8 mg, folic acid 3 mg, biotin 0.12 mg, choline chloride 0.16 mg, antioxidant 30 mg, manganese 80 mg, zinc 60 mg, iron 40 mg, copper 8 mg, iodine 0.05 mg, cobalt 0.10 mg, selenium 0.10 mg. 2Calculated metabolizable energy (g/kg, as dry matter basis) 3Methionine + Cystine Sampling procedures and biochemical analyses: Blood using the UNIVARIATE procedure of SAS (SAS, 2009). samples were collected from the wing veins of 3 All data in each dietary treatment was analyzed by randomly selected ducks in each replication. The 5 ml ANOVA using GLM procedure of SAS (SAS, 2009). samples were transferred into two plastic vials The significance of differences between the treatment containing EDTA as an anticoagulant and free-EDTA, group means for all traits was evaluated using respectively. Whole blood samples with EDTA were Duncans multiple range test, and the significant level centrifuged at 3,000×g for 10 mins at room was set at P<0.01 and P<0.05. temperature. Serum was obtained from the whole blood with free-EDTA for 10 mins at room Results temperature. The plasma and serum samples were stored at -20C until chemical analysis. The plasma uric The effects of Met sources on growth acid and triglyceride concentrations were analyzed performance are presented in Table 2. From 0 to 9 days using the PAP-method (Assay Kit, Human Gesellschaft of age, ducks fed DLM and DL-MHA significantly für Biochemica und Diagnostica GmbH, Germany). increased (P<0.01) BW, ADG, feed intake, protein Serum LDL, HDL and cholesterol were measured intake and SAA intake compared to the basal diet. using commercially available enzyme colorimetric test From 10 to 16 days of age, ducks fed DL-MHA showed kits (Roche Diagnostics GmbH, USA). the highest (P<0.01) BW and ADG when compared to DLM and basal diet, whereas ducks fed DLM and DL- Statistical analysis: The homogeneity of variances MHA improved (P<0.01) feed intake, protein intake and normal distribution of the data was confirmed and SAA intake compared to the basal diet. From 17 to

648 Kaewtapee C. and Bunchasak C. / Thai J Vet Med. 2018. 48(4): 645-653. 42 and 43 to 47 days of age, ducks fed DLM and DL- ducks fed Met (DLM or DL-MHA) significantly MHA significantly increased (P<0.01) BW and SAA increased (P<0.01) final BW, ADG, feed intake, protein intake. From 43 to 47 days of age, ducks fed DLM and intake and SAA intake compared to the basal diet. DL-MHA significantly increased (P<0.01) BW and However, there was no difference (P>0.05) in the FCR among the three groups. SAA intake. Over all periods (0 to 47 days of age), Table 2 Effect of methionine supplemented diet on growth performance of ducks from 0 to 47 d of age. Item Basal Experimental diets DL-MHA SEM P-value BW (g) 254.80B DLM 377.03A 17.90 <0.01 ADG (g/day) 22.91B 36.50A 1.99 <0.01 Feed intake (g/day) 0 to 9 day of age 2.09 <0.01 FCR 25.50B 347.95A 39.86A 0.03 0.42 Protein intake (g/day) 1.11 33.26A 1.11 0.38 <0.01 SSA intake (mg/day) 6.57B 39.80A 8.77A 2.79 <0.01 16.37B 1.20 35.87A BW (g) 8.76A 25.18 <0.01 ADG (g/day) 729.86C 35.82A 924.62A 1.84 <0.01 Feed intake (g/day) 69.58B 81.48A 2.27 <0.01 FCR 95.39B 10 to 16 days of age 111.14A 0.03 0.12 Protein intake (g/day) 868.81B 0.45 <0.01 SSA intake (mg/day) 1.37 1.38 8.74 <0.01 19.08B 72.70B 22.23A 109.40A 31.20B 93.36A 1.52 21.88A 91.89A 17 to 42 days of age BW (g) 2941.70B 3100.93A 3149.17A 31.40 <0.01 ADG (g/day) 85.14 86.52 85.79 0.68 0.75 Feed intake (g/day) 193.58 197.91 197.53 1.03 0.16 FCR 2.29 2.32 2.41 0.03 0.26 Protein intake (g/day) 35.81 36.61 36.54 0.19 0.16 SAA intake (mg/day) 82.66B 148.43A 148.15A 9.35 <0.01 43 to 47 days of age BW (g) 3182.34B 3410.14A 3375.55A 35.50 <0.01 ADG (g/day) 48.13 57.36 45.28 3.63 0.19 Feed intake (g/day) 225.26b 256.66a 234.81ab 5.43 0.03 FCR 4.82 4.46 5.28 0.30 0.57 Protein intake (g/day) 38.27b 43.61a 39.90ab 0.92 0.03 SAA intake (mg/day) 88.07B 192.50A 176.11A 14.09 <0.01 0 to 47 days of age Initial BW (g) 48.63 48.62 48.51 0.03 0.26 Final BW (g) 3182.34B 3410.14A 3375.55A 35.50 <0.01 ADG (g/d) 66.68B 71.52A 70.79A 0.76 <0.01 Feed intake (g/day) 150.65B 162.07A 159.40A 1.63 <0.01 FCR 2.26 2.27 2.25 0.04 0.83 Protein intake (g/day) 28.28B 30.32A 29.91A 0.29 <0.01 SAA intake (mg/day) 62.19B 125.73A 123.91A 8.92 <0.01 Cumulative mortality rate (%) 2.00 1.00 1.00 0.57 0.75 ADG=average daily gain; BW=body weight; FCR=feed conversion ratio; SAA intake= total sulfur amino acids intake; DLM=DL- methionine; DL-MHA=DL-methionine hydroxy analogue. a, b Means values in the same row with no common uppercase superscript differ significantly (P<0.05). A, B Means values in the same row with no common uppercase superscript differ highly significantly (P<0.01). The effects of Met sources on carcass quality greater (P<0.05) levels of blood triglyceride and HDL are presented in Table 3. Dietary supplementation of when compared to the basal diet. Furthermore, the Met (DLM or DL-MHA) significantly increased level of plasma uric acid was lower (P<0.05) with DLM (P<0.05) pectoralis major when compared to the basal supplementation than in the basal diet. diet. However, there was no significant difference (P>0.05) in pectoralis minor, drumette and leg meat Discussion among the three groups. Adding DL-MHA to the diet resulted in more (P<0.05) skin with subcutaneous fat Methionine is closely associated with lipid (breast meat area) compared to the basal diet. In metabolism as this amino acid is considered to be one contrast, supplemental Met (DLM or DL-MHA) of the dietary lipotropes (Bunchasak, 2014). significantly decreased (P<0.01) the abdominal fat Methionine deficiency can reduce feed intake, content. resulting in high fat accumulation or high activities of lipogenetic enzymes in the liver (Rosebrough and The effects of Met source on blood lipid Steele, 1985). In addition, methionine deficiency also profile and uric acid are presented in Table 4. Dietary causes an imbalance of amino acids, which results in a supplementation of DL-MHA resulted in greater large negative impact on growth performance and (P<0.05) levels of triglycerides, LDL and cholesterol in carcass quality (Bunchasak, 2014). This is in agreement the blood compared to the DLM, and in addition to

Kaewtapee C. and Bunchasak C. / Thai J Vet Med. 2018. 48(4): 645-653. 649 with the results of the present study where ducks fed a Bunchasak (2009), the relative bioefficacy of DL-MHA methionine deficient diet showed lower feed intake, compared with DLM in various animal species is from leading to higher abdominal fat accumulation and 65 to 90%. In the present study, based on 88% lower pectoralis major (breast meat) than those fed a bioefficacy (wt/wt), there was no significant difference methionine adequate diet (DLM and DL-MHA). The in growth performance between DLM and DL-MHA improvement in the performance of ducks fed Met supplementation, which was in agreement with the (DLM or DL-MHA) was in agreement with previous report of Lu and Lai (2001) and Bunchasak, (2009). studies (Xie et al., 2006; Jamroz et al., 2009; Kluge et al., Therefore, it was concluded that DL-MHA can be used 2016). The explanation is that this was due to the key as 88% bioefficacy (wt/wt) of DLM in meat type duck roles of methionine in initiating amino acid for protein at the level of the recommended requirements for the synthesis and as a principal biological methylating strain. agent in the body (Bunchasak, 2014). As reviewed by Table 3 Effect of methionine supplemented diet on carcass quality of ducks at 47 d of age (% of body weight). Item Basal Experimental Diets DL-MHA SEM P-value DLM Pectoralis major 16.35b 17.51a 17.95a 0.25 0.02 Pectoralis minor 1.66 1.80 1.78 0.05 0.49 Drumette 13.35 12.34 12.13 0.28 0.17 Leg meat 20.87 19.99 19.95 0.21 0.13 Skin with subcutaneous fat 3.90b 4.35ab 4.75a 0.13 0.02 Abdominal fat 1.23A 0.85B 1.00B 0.05 <0.01 DLM=DL-methionine; DL-MHA=DL-methionine hydroxy analogue. a, b Mean values in the same row with no common lowercase superscripts differ significantly (P<0.05). A, B Means values in the same row with no common uppercase superscript differ highly significantly (P<0.01). Table 4 Effect of methionine source on blood uric acid (mg/dL) and lipid profile (mg/dL) of ducks at 47 days of age. Item Basal Experimental diets DL-MHA SEM P-value Triglyceride 34.44b DLM 50.29a 2.62 0.02 Low-density lipoprotein 93.67a 99.63a 3.67 0.02 High-density lipoprotein 115.83b 37.88b 137.71a 4.90 0.04 75.75b 122.40ab Cholesterol 202.78ab 168.75b 216.57a 7.62 0.03 Uric acid 4.95a 3.13b 4.42ab 0.29 0.03 DLM=DL-methionine; DL-MHA=DL-methionine hydroxy analogue. a, b Mean values in the same row with no common lowercase superscripts differ significantly (P<0.05). A, B Means values in the same row with no common uppercase superscript differ highly significantly (P<0.01). Since Met is generally required for protein supplementation with DLM when an appropriated synthesis, transmethylation and transsulfuration to SAA supply is offered. cystine (Shoveller et al., 2005), dietary supplementation of Met increases the Met concentration in the body Supplementation with Met has been reported (Adeola, 2007), which supports the production of to decrease abdominal fat in broiler chickens (Jensen et contractile protein (Bikker et al., 1994). In the present al., 1989, Vieira et al., 2004, Bunchasak and Keawarun, study, Met (DLM or DL-MHA) significantly increased 2006) and meat-type duck (Xie et al., 2006), possibly breast meat (pectoralis major) but not the yield of leg because Met acts as a lipotropic agent through its role meat. This was in agreement with studies in Muscovy as a methyl donor, leading to the formation of carnitine ducks (Auvergne et al., 1991), Peking ducks (Wang et and creatine (Kalinowski et al., 2003). The function of al., 2004, Xie et al., 2006) and broiler chickens Met also plays a key role in lipogenesis and lipolysis, (Rakangtong and Bunchasak, 2011). Therefore, it is resulting in a reduction of abdominal fat accumulation stated that the breast meat of ducks is also more (Takahashi and Akiba, 1995). Notably, the amount of sensitive to SAA in the diet than that of other edible subcutaneous fat (breast meat region) obtained in the meat components. present study increased when Met was supplemented, particularly with DL-MHA. This finding suggests that There was no significant difference between there are some different effects of SAA on the the comparative effects of DLM and DL-MHA as Met regulation of adipose tissue distribution between sources on the breast meat yield and other carcass abdominal fat and subcutaneous fat. Therefore, compositions. Similarly, Bunchasak and Keawarun regional measurements of fat accumulation should be (2006) reported that supplementation with DL-MHA an important issue from the viewpoint of the carcass improved the pectoralis major of broiler chickens as quality of duck. well as DLM supplementation when 88% bioefficacy (wt/wt) was used. As a result, it is suggested that Among waterfowl, ducks have a thicker layer supplementation with DL-MHA based on 88% of subcutaneous fat to provide insulation and to bioefficacy (wt/wt) of DLM can promote protein tolerate low temperatures (Cherry and Morris, 2008). accumulation in meat-type duck similar to Consequently, skin with subcutaneous fat is greater than abdominal fat in ducks (Sakulthai, 2013). Furthermore, abdominal fat can be representative of

650 Kaewtapee C. and Bunchasak C. / Thai J Vet Med. 2018. 48(4): 645-653. body fat in broiler chickens (Bunchasak and Keawarun, Swennen et al. (2011) found that DL-MHA 2006), while the subcutaneous fat of duck has to be supplementation partially prevented the growth- taken into account when fat accumulation is concerned depressing effects of chronic heat exposure compared (Bochno et al., 2013). Interestingly, the amount of with DLM supplementation. Our results indicate that subcutaneous fat of ducks fed Met was increased, the degradation of DL-MHA and its derivatives in possibly along with the following reasons: 1) high ducks may be different from that of DLM. activity of fatty acid synthetase, which is the main lipogenic enzyme in the liver (Bunchasak et al., 1996); In conclusion, the supplementation of DL- 2) high stimulation of storage of energy as fat MHA based on 88% bioefficacy of DLM (wt/wt) at the accumulation (Smith et al., 1983); 3) high energy level of the recommended SAA requirements for the utilization due to an enhanced amino acid balance strain results in similar growth performance and (Brody, 1994); and 4) increased feed intake due to carcass quality compared to the supplementation with adding Met producing maximum growth DLM that give benefit to the production cost. performance, which results in an increase in fatty Subcutaneous fat played a key role in fat accumulation, synthesis. Likewise, DL-MHA seems to increase which can be used as fat index for meat-type duck. subcutaneous fat accumulation rather more than that However, DL-MHA seemed to increase the amount of of DLM. This phenomenon was in agreement with subcutaneous fat and the level of blood lipid profile Esteve-Garcia and Llauradó (1997) and Bunchasak and compared to DLM. The underlying mechanism of Keawarun (2006), who reported that DL-MHA some different effects between DL-MHA and DLM on induced higher fat accumulation of broiler chickens the lipid metabolism in ducks should be further compared to DLM. investigated. Previous studies (Bunchasak and Silapasorn, Acknowledgements 2005; Nukreaw et al., 2011) found that supplementation with Met increased the levels of liver We are grateful for funding provided by triglyceride and serum LDL of poultry. In the present Sumitomo Chemical Co., Ltd (Tokyo, Japan). study, supplementation with DLM resulted in a lower level of LDL, whereas supplementation of DL-MHA References led to greater levels of blood triglyceride and HDL. This finding indicates that DLM may potentially Adeola O 2007. Efficiency of methionine retention in induce hypocholesterolemia, whereas the DL-MHA ducks. Br J Nutr. 97: 478-483. may result in hypercholesterolemia in meat-type ducks. Furthermore, the differences in the levels of AOAC 2000. Official Method of Analysis. 17th ed. triglycerides, LDL and cholesterol between DLM and Gaithersberg, MD: Association of Analytical DL-MHA may be due to DLM and DL-MHA having Communities International. 2200 pp. different effects on lipoprotein synthesis. This is in agreement with studies by Esteve-Garcia and Llauradó Auvergne A, Baudonnet C and Babile R 1991. Influence (1997), who also hypothesized that the effect of DLM of protein and methionine concentrations and on fat accumulation in broiler chickens may be body size on the growth and carcass of Muscovy different from that of DL-MHA. Bunchasak et al. (2012) ducks in the finishing stage of production. Br found that DL-MHA increased the level of the estrogen Poult Sci. 32: 353-362. hormone to a greater extent than DLM. The greater estrogen hormone is responsible for an increase in the Bikker P, Verstegen MW and Bosch MW 1994. Amino hepatic synthesis of the lipo-protein, resulting in acid composition of growing pigs is affected by greater fat deposition (Scanes et al., 2004). Therefore, protein and energy intake. J Nutr. 124: 1961-1969. Met source may have different ways of influencing the function of hormones, which are involved in lipid Bochno R, Murawska D and Michalik D 2013. A metabolism. modified skin slice with subcutaneous fat and carcass weight without this slice as indicators of Uric acid is a major end product of protein total lean meat and fat content in poultry catabolism in poultry and its concentration in the blood carcasses. Afr J Agric Res. 8: 5859-5863. is also in response to the contents of protein and amino acids in the diet (Miles and Featherston, 1974, Brody T 1994. Nutritional Biochemistry. San Diego, Donsbough et al., 2010). The present study found that CA, USA: Academic Press.658 pp. supplementing DLM to meet the requirements of ducks reduced plasma uric acid due to an Bunchasak C, Tanaka K, Ohtani S and Collado CM improvement in the amino acids balance (Xie et al., 1996. Effect of met+cys supplementation to a low- 2004). However, the level of uric acid in ducks fed a protein diet on the growth performance and fat diet supplemented with DL-MHA did not differ from accumulation of broiler chicks at starter period. that in ducks fed the basal diet. On the other hand, uric Anim Sci Technol (Jpn). 67: 956-966. acid is an antioxidant agent and poultry can use it to scavenge free radical products from the metabolic Bunchasak C, Santoso U, Tanaka K, Ohtani S and process that results in prevention of DNA damage and Collado CM 1997. The effect of supplementing lipid oxidation (Stinefelt et al., 2005). In broiler methionine plus cystine to a low-protein diet on chickens, Yodseranee and Bunchasak (2012) reported the growth performance and fat accumulation of that the levels of blood uric acid and taurine were growing chicks. Asian-Australas J Anim Sci. 10: increased by DL-MHA supplementation. Moreover, 185-191. Bunchasak C and Silapasorn T 2005. Effects of adding methionine in low-protein diet on production performance, reproductive organs and chemical liver composition of laying hens under tropical conditions. Int J Poult Sci. 4: 301-308.

Kaewtapee C. and Bunchasak C. / Thai J Vet Med. 2018. 48(4): 645-653. 651 Bunchasak C and Keawarun N 2006. Effect of chicken and duck embryos. Toxicol Appl methionine hydroxy analog‐free acid on growth Pharmacol. 172: 241-248. performance and chemical composition of liver of Kalinowski A, Moran Jr ET and Wyatt CL 2003. broiler chicks fed a corn-soybean based diet from Methionine and cystine requirements of slow- 0 to 6 weeks of age. Animal Sci J. 77: 95-102. and fast-feathering broiler males from three to six weeks of age. Poult Sci. 82: 1428-1437. Bunchasak C 2009. Role of dietary methionine in Kluge H, Gessner DK, Herzog E and Eder K 2016. poultry production. J Poult Sci. 46: 169-179. Efficacy of DL-methionine hydroxy analogue- free acid in comparison to DL-methionine in Bunchasak C, Ratchadapornvanitch Y and Thiengtham growing male white Pekin ducks. Poult Sci. 95: J 2012. Comparative effects of supplemental DL- 590-594. 2-hydroxy-4-[methylthio] butanoic acid and DL- Llames CR.and Fontaine J 1994. Determination of methionine in diet on egg production and quality amino acids in feeds: Collaborative study. J Assoc in laying hens. J Poult Sci. 49:260-267. Off Anal Chem. 77: 1362-1402. Lu JJ and Lai FJ 2001. The effects of DL-methionine and Bunchasak C 2014. Comparative bioefficacy of DL-methionine hydroxyl analogue on growth supplemental DL-2-hydroxy-4-[methyl] butanoic performance protein digestive enzymes activities acid and DL-methionine on absorption, in Pekin ducks. J Chin Soc Anim Sci. 30: 143-153. conversion and production performance. Miles RD and Feathersto WR 1974. Uric acid excretion Bangkok, Thailand: Danex Intercorporation Co., as an indicator of the amino acid requirement of Ltd. 157 pp. Chicks. Proc Soc Exp Biol Med. 145: 686-689. Mooney RA and Lane MD 1981. Formation and Chen F, Noll SL, Waibel PE and Hawkins DM 1993. turnover of triglyceride-rich vesicles in the chick Effect of folate, vitamin B12 and choline liver cell. Effects of cAMP and carnitine on supplementation on turkey breeder performance. triglyceride mobilization and conversion to Poult Sci. 72(Suppl. 1): 73 (Abstr.). ketones. J Biol Chem. 256: 11724-11733. NRC 1994. Nutrient Requirements of Poultry. 9th rev. Cherry Valley Farms. 2004. SM3 commercial ed. Washington, DC: National Academy Press. management manual. UK: Laceby. 42 pp. 176 pp. Nukreaw R, Bunchasak C, Markvichitr K, Choothesa Cherry P and Morris TR 2008. Domestic duck A, Prasanpanich S, Loongyai W 2011. Effects of production: science and practice. Wallingford, methionine supplementation in low-protein diets Oxfordshire: CABI. 239 pp. and subsequent re-feeding on growth performance, liver and serum lipid profile, body Commission Directive 2000. Establishing community composition and carcass quality of broiler methods of analysis for the determination of chickens at 42 days of age. J Poult Sci. 48: 229-238. vitamin A, vitamin E and tryptophan in Rakangtong C and Bunchasak C 2011. Effect of total feedingstuffs. Annex part C. Determination of sulfur amino acids in corn–cassava–soybean diets tryptophan. Off J Eur Union. L 174: 32-50. on growth performance, carcass yield and blood chemical profile of male broiler chickens from 1 Donsbough AL, Powell S, Waguespack A, Bidner TD to 42 days of age. Animal Prod Sci. 51: 198-203. and Southern LL 2010. Uric acid, urea and Ross 308. 2014. Broiler nutrition specifications. ammonia concentrations in serum and uric and Aviagen, Midlothian, Scotland. 10 pp. uric acid concentration in excreta as indicators of Rosebrough RW and Steele NC 1985. Energy and amino acid utilization in diets for broilers. Poult protein relationships in the broiler. 1. Effect of Sci. 89: 287-294. protein levels and feeding regimens on growth, body composition, and in vitro lipogenesis of Esteve‐Garcia E and Llauradó L 1997. Performance, breast meat yield and abdominal fat deposition of broiler chicks. Poult Sci. 64: 119-126. male broiler chickens fed diets supplemented SAS 2009. SAS/STAT. User’s Guide, Version 9.2. Cary, with DL‐methionine or DL‐methionine hydroxy analogue free acid. Br Poult Sci. 38: 397-404. NC: SAS Inst. Inc. 7869 pp. Sakulthai S 2013. Effect of DL-methionine and liquid Grimaud Freres 2010. Rearing guide roasting pecking ducks. La Corbière, Roussay: Grimaud Freres. 19 DL-methionine hydroxy analog free acid on pp. growth performance, carcass quality and meat quality of broiler chickens and ducks. Master Hermier D 1997. Lipoprotein metabolism and fattening Thesis. Bangkok, Thailand: Kasetsart University. in poultry. J Nutr. 127: 805S-808S. 104 pp. Scanes C, Brant G, Ensminge ME 2004. Poultry Science. Jamroz D, Wiliczkiewicz A, Lemme A, Orda J, New Jersey: Pearson Prentice. Skorupińska J and Wertelecki T 2009. Effect of Shoveller AK, Stoll B, Ball RO and Burrin DG 2005. increased methionine level on performance and Nutritional and functional importance of apparent ileal digestibility of amino acids in intestinal sulfur amino acid metabolism. J Nutr. ducks. J Anim Physiol Anim Nutr. 93: 622-630. 135: 1609-1612. Smith JT, Acuff RV, Bittle JB and Gilbert ML 1983. A Jensen LS Wyatt CL and Francer BI 1989. Sulphur metabolic comparison of cysteine and methionine amino acid requirement of broiler chickens from 3 to 6 weeks of age. Poult Sci. 68: 163-168. Jianlin S, Kersey J, Fritts C and Waldroup P 2004. An evaluation of the interaction of lysine and methionine in diets for growing broiler. Int J Poult Sci. 3: 51-60. Jin X, Kennedy SW, Di Muccio T and Moon TW 2001. Role of oxidative stress and antioxidant defense in 3, 3′, 4, 4′, 5-pentachlorobiphenyl-induced toxicity and species-differential sensitivity in

6502 Kaewtapee C. and Bunchasak C. / Thai J Vet Med. 2018. 48(4): 645-653. supplements in the diet of a rat. J Nutr. 113: 222- 227. Stinefelt B, Leonard SS, Blemings KP, Shi X and Klandorf H 2005. Free radical scavenging, DNA protection, and inhibition of lipid peroxidation mediated by uric acid. Ann Clin Lab Sci. 35: 37- 45. Swennen,Q, Geraert P, Mercier Y, Everaert N, Stinckens A, Willemsen H, Li Y, Decuypere E and Buyse J 2011. Effects of dietary protein content and 2-hydroxy-4-methylthiobutanoic acid or dl- methionine supplementation on performance and oxidative status of broiler chickens. Br J Nutr. 106: 1845-1854. Takahashi K and Akiba Y 1995. Effect of methionine supplementation on lipogenesis and lipolysis in broiler chicks. Jpn Poult Sci. 32: 99-106. Tillman PB and Pesti GM 1986. The response of male broiler chicks to a corn-soy diet supplemented with L-methionine, L-cystine, choline, sulfate, and vitamin B12. Poult Sci. 65: 1741-1748. Vieira SL, Lemme A, Goldenberg DB and Brugalli I 2004. Responses of growing broilers to diets with increased sulfur amino acids to lysine ratios at two dietary protein levels. Poult Sci. 83: 1307- 1313. Wang YZ, Xu ZR and Feng J 2004. The effect of betaine and dl-methionine on growth performance and carcass characteristics in meat ducks. Anim Feed Sci Technol. 116: 151-159. Wong PC, Vohra P and Kratzer FH 1977. The folacin requirement of broiler chicks and quail (Coturnixcoturnix japonica). Poult Sci. 56: 1852- 1860. Xie M, Hou SS, Huang W, Zhao L, Yu JY, Li WY and Wu YY 2004. Interrelationship between methionine and cystine of early Peking ducklings. Poult Sci. 83: 1703-1708. Xie M, Hou S and Huang W 2006. Methionine requirements of male white Peking ducks from twenty-one to forty-nine days of age. Poult Sci. 85: 743-746. Yodseranee R and Bunchasak C 2012. Effects of dietary methionine source on productive performance, blood chemical, and hematological profiles in broiler chickens under tropical conditions. Trop Anim Health Prod. 44: 1957-1963.

Kaewtapee C. and Bunchasak C. / Thai J Vet Med. 2018. 48(4): 645-653. 6531 บทคดั ย่อ การเปรยี บเทียบผลการเสรมิ ดแี อล-เมทไธโอนนี และเมทไธโอนีนไฮดรอกซีอะนาลอกในอาหารต่อ สมรรถภาพการผลิต การสะสมไขมนั และไขมันในเลือดของเป็ดเนือ้ ชาญวทิ ย์ แกว้ ตาปี1* ชยั ภูมิ บัญชาศกั ด์ิ1 วัตถปุ ระสงค์ของการศึกษาน้เี พื่อเปรยี บเทียบผลการเสริมดีแอล-เมทไธโอนนี (DLM) และเมทไธโอนีนไฮดรอกซอี ะนาลอก (DL- MHA) ต่อการเจริญเติบโต คณุ ภาพซากและไขมนั ในเลอื ดของเปด็ เนือ้ โดยใชเ้ ปด็ เพศผ้พู ันธุ์เชอรร์ ่ีวลั เลย์จานวน 300 ตัว แบง่ ออกเป็น 3 กลุ่มๆ ละ 4 ซา้ ๆ 25 ตัว เป็ดแตล่ ะกลุ่มได้รับอาหารดงั ตอ่ ไปน้ี 1) อาหารพ้ืนฐาน 2) อาหารพ้ืนฐาน + DLM และ 3) อาหารพ้นื ฐาน + DL-MHA จาก การศึกษาพบวา่ อาหารท่เี สริม DLM และ DL-MHA มีผลเพ่มิ น้าหนักตัว (P<0.01) ปรมิ าณอาหารท่ีกนิ (P<0.01) และเน้ือหน้าอก (P<0.05) แตม่ ผี ลทาให้ไขมันชอ่ งทอ้ งลดลง (P<0.01) เมือ่ เปรียบเทียบกับกลมุ่ ควบคมุ การเสรมิ DL-MHA มผี ลเพ่ิมระดบั ไตรกลเี ซอร์ไรด์ (P<0.05) Low-density lipoprotein (P<0.05) และคลอเลสเตอรอล (P<0.05) ในเลอื ดเม่อื เปรียบเทียบกับกลมุ่ ทีเ่ สริม DLM อีกทงั้ มผี ลเพมิ่ ระดบั ไตร กลเี ซอร์ไรด์ (P<0.05) และ High-density lipoprotein (P<0.05) เมื่อเปรยี บเทยี บกับกลมุ่ ควบคุม นอกจากน้ีการเสรมิ DLM มผี ลลดระดบั กรดยรู กิ (P<0.05) ในพลาสมาเมื่อเทียบกับกลมุ่ ควบคมุ จงึ สรุปได้วา่ การเสริม DL-MHA สามารถสง่ ผลต่อการเจริญเติบโตไดเ้ ช่นเดียวกับการ เสรมิ DLM แตก่ ล่มุ ท่ีเสรมิ DL-MHA จะมีการสะสมไขมันในร่างกายและไขมันในเลอื ดสูงกว่ากลมุ่ ที่เสริม DLM อยา่ งไรกต็ ามกลไกทแี่ ตกตา่ ง กนั จากการเสรมิ DL-MHA และ DLM ต่อเมแทบอลซิ มึ ของไขมัน ควรมกี ารศึกษาตอ่ ไป คาสาคัญ: ดแี อล-เมทไธโอนีน เมทไธโอนนี ไฮดรอกซอี ะนาลอก การสะสมไขมนั เปด็ เน้ือ สมรรถภาพการผลิต 1ภาควชิ าสัตวบาล คณะเกษตร มหาวทิ ยาลัยเกษตรศาสตร์ 50 ถนนงามวงศ์วาน แขวงลาดยาว เขตจตุจกั ร กรุงเทพฯ 10900 *ผู้รับผดิ ชอบบทความ E-mail: [email protected]



Original Article Survival time and prognosis factors in hypertrophic cardiomyopathy cats with congestive heart failure Lalida Tantisuwat1 Panduangjai Puangampai1 Pornchanok Panpakdee1 Thitinan Tangarsasilp1 Sirilak Disatian Surachetpong2* Abstract Hypertrophic cardiomyopathy (HCM) is a common myocardial disease in cats. Nowadays, there is no study determining the survival time of cats affected with HCM in Thailand. This study aimed to evaluate the survival time in HCM cats with congestive heart failure (CHF) in Thailand and to determine the factors associating with survival time. A retrospective study was performed in 47 HCM cats to analyze survival time and the association of survival time and factors including sex, age, breeds, weight, physical findings, the presence of complications, medication and echocardiographic findings. The data was analyzed by the Log-Rank test and displayed with the Kaplan-Meier curve. The results reveal that the median survival time of HCM cats with CHF was 283 days and factors associating with a lesser survival time were the presence of aortic thromboembolism (ATE), a decrease in left ventricular internal dimension at end-diastole (LVIDd) < 10.8 mm and the use of other cardiovascular drugs besides the combination of angiotensin converting enzyme inhibitors (ACEi) and furosemide. The data of this study is valuable for prognosis and for the future management of HCM cats with CHF. Keywords: cat, congestive heart failure, hypertrophic cardiomyopathy, survival time 16th year student, academic year 2017, Faculty of Veterinary Science, Chulalongkorn University, 39 Henri-Dunant Rd., Pathumwan, Bangkok, 10330 Thailand 2Department of Veterinary Medicine, Faculty of Veterinary Science, Chulalongkorn University, 39 Henri-Dunant Rd., Pathumwan, Bangkok, 10330 Thailand *Correspondence: [email protected] Thai J Vet Med. 2018. 48(4): 655-662.

656 Tantisuwat L. et al. / Thai J Vet Med. 2018. 48(4): 655-662. Introduction confirmed, these cats were excluded from the study. Data of cats dying from other causes than cardiac death Hypertrophic cardiomyopathy (HCM), the was not included. most common heart disease in cats, is caused by an abnormality of myocardium related to myocardial All cats had echocardiography performed by dysfunction. Hypertophic cardiomyopathy is one investigator. The diagnostic criteria for HCM commonly found in Thailand but its prevalence has not consisted of a thickened left ventricular wall (≥6 mm) been published. This disease is usually diagnosed in and a decreased left ventricular chamber size. the middle-age of cats (median age 6.2 years with a Echocardiographic data including types of range of 0.75 to 17.4 years) (Payne et al., 2013). Male hypertrophy (symmetry/asymmetry), left ventricular cats are more prone to the disease (Payne et al., 2013). internal dimension at end-diastole (LVIDd), fractional Hypertrophic cardiomyopathy is commonly found in shortening, the presence of spontaneous contrast in the domestic short hair (DSH) and purebred cats such as left atrium, left atrial dimension at end-diastole (LA), Ragdoll, Maine Coon, Himalayan, Burmese, Sphynx, the ratio of left atrial dimension to the aortic annulus Persian, and Devon Rex. dimension (LA:Ao) and the ratio of peak velocity of early diastolic to late diastolic transmitral flow of Hypertrophic cardiomyopathy primarily mitral valve (E:A) were recorded (Boon, 2017). affects the left ventricular myocardium and can cause left-sided congestive heart failure (CHF) in the late Data analysis was performed by a commercial stage of the disease. Intracardiac thrombi and aortic statistical program (IBM SPSS 22, USA). Descriptive thromboembolism (ATE) may develop in HCM cats statistical analysis was used to identify the population with a severely enlarged atrium (White, 2015). characteristics. The univariable Cox-regression was analyzed for the hazard of death. The survival time There have been several studies reporting the analysis was performed by the Kaplan-Meier survival survival time in HCM cats (Atkins et al., 1992; Rush et curve. The difference of the survival curves was al., 2002; Payne et al., 2010). However, there has been compared by the Log-rank test. P < 0.05 was no report in Thailand. Previous studies indicate that considered to be of statistical significance. The survival the prognosis of cats with HCM is highly variable time was presented as median and 95% confident depending on variations in diagnosis and treatment interval (CI). The data from live cats at the end of the protocols, clinician opinions, the financial constraints study was censored. of the owner and owner compliance. Payne et al. (2013) showed that cats presented with an extreme left Results ventricular hypertrophy (≥9 mm), left ventricular fractional shortening ≤30%, and decreasing in the left Data from 47 HCM cats with CHF was atrium to %fractional shortening had a poor retrieved. The median survival time of 47 HCM cats prognosis. with CHF was 283; 95% CI 1-596 days. As mentioned previously, HCM is a life- Breeds of HCM cats in this study included threatening disease in cats. The purpose of the study DSH (48%, 23/47), Persian (40.4%, 19/47), American reporting here is to determine the survival time and Shorthair (4.26%, 2/47), Maine Coon (2.12%, 1/47), prognostic indicators of HCM cats with CHF in Scottish Fold (2.12%, 1/47), and Exotic Shorthair (2.12%, Thailand. The results of this study will be valuable 1/47). There were 51.1% males (24/47), seventeen intact information for veterinarians to provide a prognosis to and seven neutered male cats, and 48.9% female cat owners in Thailand. (23/47), six intact and seventeen spayed female cats. The descriptive data of cats recruited to the study is Materials and Methods presented in Table 1. Twenty-six cats had cardiomegaly with pulmonary edema. Thirteen cats A retrospective study was performed. had pleural effusion, and four cats had pleural effusion Electronic medical records of the Small Animal and ascites. Four cats had both pulmonary edema and Teaching Hospital, Faculty of Veterinary Science, pleural effusion. Twenty-five cats died during the Chulalongkorn University from March 2014 to March study period and twenty-two cats were alive at the end 2017 were retrieved to identify cats diagnosed with of the study. HCM. The day of the echocardiographic examination was defined as the first day of a visit. Information on The result from the Cox-regression analysis is cats including sex, age, breed, body weight, heart summarized in Table 2. Hypertrophic cardiomyopathy sound, radiographic findings, echocardiographic cats with the presence of ATE and a decrease in LVIDd findings, systolic blood pressure, manifestation of had an increased hazard of death compare to those CHF, the presence of arterial thromboembolism (ATE), without these conditions (Table 2). In addition, HCM the presence of concurrent chronic kidney disease cats treated with ACEi and furosemide had a (CKD), cardiac medical treatments, duration of decreased hazard of death compared with those treatment, live status, and date of death was recorded. treated with other cardiovascular drugs (Table 2). The criteria for CHF diagnosis were clinical signs including respiratory distress or rapid breathing and A comparison of survival times of HCM cats radiographic findings including pulmonary edema, depending on different categories is summarized in pleural effusion, and/or ascites. In the case of cats that Table 3. The presence of ATE, a decrease of LVIDd, and did not have a death-alive-status or date of death at the the use of ACEi and furosemide affected the survival end of the study in March 2017, the death-alive-status time of HCM cats with CHF (Table 3). The survival was confirmed by telephone interviewing with the time was not different when analyzed with other owners. If the death-alive status could not be parameters including age, breed, sex, heart sound, systolic blood pressure, % fractional shortening, types

Tantisuwat L. et al. / Thai J Vet Med. 2018. 48(4): 655-662. 657 of hypertrophy, LA, LA:Ao, E:A, and the presence of furosemide including diltiazem, pimobendan, or spontaneous contrast. atenolol (p = 0.019) (Figure 3, Table 5). The survival time between HCM cats with There were thirteen cats presented with CKD. ATE and without ATE was distinctly different (p < The median survival time of 17 cats with CKD (339; 95% CI 60-617 days) and without CKD (230; 95% CI 96- 0.0001) (Figure 1, Table 5). Cats with LVIDd <10.8 mm 363 days) (p=0.657) was no different. A sub-analysis had a shorter survival time than those with LVIDd from data of HCM cats without CKD was performed. >10.8 mm (p = 0.009) (Figure 2, Table 5). According to The comparison of survival time of HCM cats without the treatment protocol, cats treated with ACEi and CKD depending on different categories is summarized furosemide had a longer survival time than cats treated in Tables 4 and 5. with other cardiovascular drugs besides ACEi and Table 1 Descriptive data of hypertrophic cardiomyopathy cats recruited in the study Category Median Interquartile range Age (years) 8 3.8-12 Weight (kg) 3.8 3.0-4.8 Blood pressure (mmHg) 140 115-170 Echocardiography Fractional shortening (%) 58.13 47.19-67.47 LVIDd (mm) 13.4 12.2-15.6 LA (mm) 15.6 12.6-17.4 LA:Ao 1.8 1.35-2.31 E:A 1.18 0.86-2.24 LVIDd, left ventricular internal dimension at end-diastole; LA, left atrium; LA:Ao, the ratio of left atrial dimension to the aortic annulus dimension; E:A, the ratio of peak velocity of early diastolic to late diastolic transmitral flow of mitral valve Table 2 The result of the Cox-regression analysis for the hazard of death Parameters Category Hazard 95% CI P value ratio Age (years) <7 and >7 0.284-1.559 0.348 Breed DSH and others 0.665 0.635-3.09 0.405 1.4 0.436-2.168 0.944 Sex Persian and others 0.972 0.529-2.696 0.668 DSH and persian 1.195 0.259-1.269 0.170 Male and female 0.573 Heart sound Normal and abnormal 1.430 0.622-3.287 0.400 Blood pressure (mmHg) <160 and >160 1.262 0.443-3.591 0.663 ATE Presence and absence 5.169 1.932-13.834 0.001 CKD Presence and absence 1.230 0.491-3.085 0.658 Echocardiography <40, 40-60 and >60 0.872 0.248-3.069 0.831 Fractional shortening (%) 1.145 0.520-2.523 0.737 Types of hypertrophy Asymmetry and symmetry 3.257 1.274-8.327 0.014 LVIDd (mm) 1.277 0.579-2.816 0.544 LA (mm) <10.8 and >10.8 0.614 0.280-1.347 0.224 LA:Ao 0.304 0.080-1.150 0.080 E:A <16 and >16 1.027 0.407-2.594 0.955 Spontaneous contrast <2 and >2 <1: impaired 1-2: pseudonormalized >2: restrictive Presence and absence Medication ACEi and Furosemide With and without 0.377 0.161-0.881 0.024 ACEi, angiotension converting enzyme inhibitor; ATE, aortic thromboembolism; CI, confident interval; CKD, chronic kidney disease; DSH, domestic short hair; LVIDd, left ventricular internal dimension at end-diastole; LA, left atrium; LA:Ao, the ratio of left atrial dimension to the aortic annulus dimension; E:A, the ratio of peak velocity of early diastolic to late diastolic transmitral flow of mitral valve