key: cord-0998266-6gab178z
authors: Roy, RD; Edens, FW; Parkhurst, CR; Qureshi, MA; Havenstein, GB
title: Influence of a propionic acid feed additive on performance of turkey poults with experimentally induced poult enteritis and mortality syndrome
date: 2002-07-01
journal: Poultry Science
DOI: 10.1093/ps/81.7.951
sha: 475c30b92a6a6da5b490449ac9f4329ac6b710f7
doc_id: 998266
cord_uid: 6gab178z

Abstract Poult enteritis and mortality syndrome (PEMS) has multiple etiological agents associated with its occurrence, including two viruses and at least three Escherichia coli isolates. Myco Curb (MC) contains organic acids and is used as a feed additive to inhibit growth of many bacteria and toxin-producing molds but not viruses. Studies evaluating the influence of MC on BW, feed conversion, and mortality indicate that turkey poults tolerate MC at 1.25% but not 2.50%, but higher MC content in feed provides greater suppression of growth of bacterial isolates commonly associated with PEMS. In two PEMS experiments, 1.25% MC was blended into poult starter feed and was maintained in the feed for the duration of the 3-wk experiments. In these experiments, 1-d-old commercial poults were placed into battery brooders and were given turkey starter feed and water ad libitum. At 6 d posthatch, PEMS-designated poults were given a 1-mL oral gavage of a 10% suspension of feces from PEMS-infected poults. BW depression due to PEMS was not alleviated by MC, although there was less variation in mean BW of the MC-fed poults, and there was a highly significant reduction in mortality (68% in PEMS-exposed with MC vs. 32.5% in PEMS-exposed without MC). The reduction in mortality in the MC-fed poults was attributed to decreased bacterial content of the gut and to maintenance of packed cell volume and hemoglobin content. It was concluded that MC might be a potential nutritional intervention during PEMS.

Poult enteritis and mortality syndrome (PEMS) is a term that describes potentially lethal enteritis of young turkeys (Barnes et al., 1996) . It is caused by agents such as bacteria and viruses that irritate and injure the intestines (Edens et al., 1997b,c,d; Guy et al., 2000; Heggen et al., 2000; Qureshi et al., 2000; Shultz-Cherry et al., 2000 , 2001 Yu et al., 2000a,b) .

Immune dysfunction associated with PEMS (Qureshi et al., 1997; Heggen et al., 2000) increases susceptibility to secondary bacterial infection that ultimately will kill poults (Edens et al., 1997b,c,d) . Edens et al. (1997b,c,d) consistently isolated from moribund, PEMS-infected poults two atypical strains of Escherichia coli (colony type 1 and colony type 2). Furthermore, Edens et al. (1997b,c,d) observed that the E. coli strains involved in PEMS were resistant to most of the antibiotics commonly used by To whom correspondence should be addressed: fwedens@ mindspring.com. 951 old commercial poults were placed into battery brooders and were given turkey starter feed and water ad libitum. At 6 d posthatch, PEMS-designated poults were given a 1-mL oral gavage of a 10% suspension of feces from PEMSinfected poults. BW depression due to PEMS was not alleviated by MC, although there was less variation in mean BW of the MC-fed poults, and there was a highly significant reduction in mortality (68% in PEMS-exposed with MC vs. 32.5% in PEMS-exposed without MC). The reduction in mortality in the MC-fed poults was attributed to decreased bacterial content of the gut and to maintenance of packed cell volume and hemoglobin content. It was concluded that MC might be a potential nutritional intervention during PEMS. the turkey industry. Therefore, the observation that there were some apparently atypical E. coli isolates involved in PEMS represented a significant breakthrough in the diagnosis and control of PEMS.

Because PEMS is a major economic threat to the turkey production sector of the U.S. poultry industry, it is important to explore potential interventions that may ameliorate its severity. The development of PEMS appears to be influenced by Escherichia coli, Salmonellae, and other Enterobacteriaceae (Edens et al., 1997b,c,d) . There are at least two atypical E. coli along with numerous other untyped E. coli involved in the PEMS problem (Edens et al., 1997b,c,d) . Additionally there are at least 14 isolates of Salmonellae and at least one atypical Salmonella that are commonly associated with PEMS, and most of these isolates show a broad range of resistance to antibiotics used by the turkey industry (Edens et al., 1997b,c,d) .

There may be interactions among different bacterial pathogens that contribute to PEMS and exacerbate its severity (Edens et al., 1997b,c,d) . A single population of pathogenic microbes often can be competitively regulated or excluded by other microbes in the intestinal tract with little negative influence on the performance of the poult (Edens et al., 1997a) , but combinations of microbes or feed molds (toxins) may allow proliferation of normally innocuous bacteria that will make the poult more susceptible to disease (Edens et al., 1997a) . Environmental conditions, nutrition, and chemotherapeutic chemicals can influence the control of microbial populations and the effects of molds and their toxins in poultry species such as the turkey poult (Edens, 1996; Edens and Doerfler, 1998; .

Therefore, intervention research that addresses the potential use of nonantibiotic substances as a means for amelioration of the detrimental effects of PEMS is important. Myco Curb (MC), a mold inhibitor for processed feed ingredients and animal feeds, contains organic acids (primarily propionic acid) that are effective against various bacteria and fungi (Kemin Industries, 1986 Pumfrey and Nelson, 1991; Hume et al., 1993; Larson et al., 1993; Huff et al., 1994) . Anecdotal information from a turkey producer in Western North Carolina indicated that MC (0.625%) reduced mortality in flocks of poults that were suspected to have developed PEMS (without MC, 12.47%; with MC, 5.52%). Therefore, a product such as MC appeared to have potential for immediate use as a nutritional intervention in PEMS.

This project was approved and conducted under the supervision of the North Carolina State University Animal Care and Use Committee, which has adopted Animal Care and Use Guidelines governing all animal use in experimental procedures.

British United Turkey poults were obtained and transported from a commercial hatchery to North Carolina State University. These 1-d-old poults were wing-banded, weighed, and placed, in groups of 10, into pens in Petersime 3 heated battery brooders where water and turkey starter feed (North Carolina Agricultural Research Service; metabolizable energy: 2,915 kcal/kg; crude protein: 28.13%) with or without MC were provided ad libitum. The poults were not subjected to hatchery services such as beak or toe trimming, antibiotic administration, or vaccinations. Continuous lighting was provided by incandescent lamps in the ceiling of each room. Two experiments were conducted. The first was a dose titration study to determine an adequate level of MC that would maximize the growth and feed conversion of male and female poults, and the second was conducted to determine the influence of MC on performance of PEMS-infected male poults. At the time of placement for Experiment 2, two groups of poults (controls and those designated for PEMS-exposure) were assigned to separate but identical controlled-environment isolation rooms.

Ambient temperature for brooding was maintained by room air conditioning using a thermostatically controlled hot/cold water heat exchange system mediated by a forced draft. Initial room brooding temperature for the control and PEMS rooms was set at 34 C, and this temperature was decreased 3 C in each room at 7, 14, and 21 d of brooding. Humidity in the experimental rooms was not controlled and varied from 47 to 63% relative humidity.

The dose titration experiment used a 2 × 4 × 5 factorially arranged completely randomized statistical design. There were 50 male and 50 female poults in each of four different feed additive treatments. MC 4 was blended in dry mash diets to provide a dietary concentration of 0, 0.625, 1.25, or 2.50%. There were five replicate pens assigned for each sex and MC dietary treatment. A total of 50 pens with 10 poults per pen were involved. Thus, 500 poults were included in the first experiment, which was conducted over 21 d. Based on the results from Experiment 1, a 1.25% MC addition rate was chosen for use in Experiment 2. The second experiment used a factorially arranged 2 (control vs. PEMS) × 2 (0 vs. 1.25% MC) completely randomized experimental design. The MC was blended in feed as described above. Two trials were conducted utilizing this design involving 300 poults per trial. Poults in groups of 10 each were placed into heated battery brooders. Five replicate pens were used for each treatment control group with or without MC, and there were 10 replicate pens for each PEMS treatment groups with or without MC.

At 6 d posthatch, each poult in the PEMS-exposed groups was given a 1-mL oral gavage of a 10% suspension of feces from Coronavirus-negative PEMS-positive poults maintained at the College of Veterinary Medicine at North Carolina State University. This procedure has been documented to induce in experimental conditions a disease state that was equivalent to field cases of PEMS (Edens et al., 1997b,c) .

Body weights were taken at placement and at weekly intervals through 21 d of age. Birds that died during each of Weeks 1, 2, or 3 were weighed, and their weight gain was included in calculations of weekly feed conversion ratios. Blood samples were collected on a weekly interval for determination of packed cell volume (PCV) and hemoglobin (Hgb) concentration (Phelps et al., 1987) and the Within a column, means with unlike superscripts differ significantly (P < 0.05). PEMS-exposed poults with and without MC were compared with the unexposed control poults with and without MC.

In earlier studies, we had determined that the great majority of the bacterial isolates from PEMS-infected poults were Enterobacteriaceae (Edens et al., 1997a,b,c,d) . Therefore, in Experiment 1, it was decided to monitor Enterobacteriaceae in cecal samples collected from control poults with and without MC in their diets. Poults selected for determination of Enterobacteriaceae were 14 d of age, and if they had been PEMS-infected, they would be entering into a phase representing the most severe period of diarrhea that would culminate in mortality. Ceca, weighing about 1 g, were collected and placed in sterile stomacher bags containing 9 mL of tetrathionate 5 and thoroughly emulsified. The samples were streaked onto differential media with minimal selectivity for Enterobacteriaceae, such as MacConkey's agar 5 with 0.15% bile salts, XLT4 agar, 5 tryptose soy agar 5 with 5% sheep blood, and Levine eosin methylene blue (EMB) agar. 5 All cultures were grown at 37 C for 24 h.

Because colony appearance can be altered by close association with other bacteria, 20 representatives of each selected colony type from each of the five cecal samples per treatment were carefully collected by a straight wire directed at the center of a colony without touching the agar surface. Relative plate growth (4+: extensive/over growth/too many to count, 3+: extensive growth/colonies can be enumerated, 2+: moderate growth/easily enumerated colonies; 1+: only a few colonies per plate; <1+: sporadic appearance of bacterial colonies on the plate) of the different selected colonies was determined by examination of the plates with a magnifier associated with the colony counter. These isolates were then grown anaerobically 24 h at 37 C in brain heart infusion broth 5 (BHI). Differentiation of tribes and sometimes genus of selected Enterobacteriaceae was accomplished by developing biochemical profiles using the BBL Enterotube II System 6 and the API 20E System 7 according to manufacturer's 5 Difco Laboratories, Detroit, MI. protocols following procedures outlined by Ewing (1986) as reported by Edens et al. (1997c,d) .

All BW, feed conversion, mortality, and blood component data were subjected to analysis of variance using the general linear models procedure of SAS software (SAS Institute, 1996) . Statements of significance are based on P ≤ 0.05 as a minimum level of significance.

The data for the influence of various doses of MC on performance of male and female turkey poults are presented in Table 1 . Male and female poults had a significantly lower BW when they were given 2.5% MC in their feed (Table 1) . Furthermore, feed conversion was increased with the 2.5% MC supplementation. However, at 14 d of age, when the poults would be developing severe PEMS-associated diarrhea, the relative growth of selected bacterial species commonly found in association with PEMS was decreased with increasing dietary concentrations of MC (Table 2) . Based on the effects of MC on common enteric bacterial species found in conjunction with PEMS (Table 2) , as well as feed conversion ratios and male and female poult BW in a dose titration study (Table 1) , it was ascertained that the 1.25% MC level would be added to the feed in a second experiment focused on the influence of MC on the development of PEMS.

In the second experiment, at 6 d postplacement, the poults in the PEMS-designated isolation room were challenged via oral gavage. At Day 7, only 24 h after challenge, the PEMS-exposed poults had slightly depressed BW compared to the nonexposed control and MC-treated poults (Figure 1 ). By 14 d of age, the influence of PEMS exposure on BW was clearly evident and significant (Figure 1) . Although the surviving poults were severely stunted at 21 d of age, they were beginning to grow again.

The mean daily mortality indicated that the MC feed additive did not prevent the disease from occurring in the PEMS-exposed poults (Figure 2 ). However, MC did delay the onset of the initial mortality spike associated with the disease (Figures 2 and 3) and reduced the cumulative mortality by more than one-half (68 vs. 32.5%; Figure 3) . Indicates relative bacterial growth on plates; 4+: extensive/over growth/too many to count, 3+: extensive growth/colonies can be enumerated, 2+: moderate growth/easily enumerated colonies; 1+: only a few colonies per plate; <1+: sporadic appearance of bacterial colonies on the plate. 3 These were the only types found in this treatment group.

The PCV of nonexposed control and MC-treated poults did not differ between groups (ranging between 32 and 35%) throughout the experiment (Figure 4) . However, the PEMS-exposed poults experienced a rapid increase in PCV that was elevated significantly within 2 d after exposure (Figure 4) . The PCV in the PEMS-exposed groups peaked at 45% at 9 d after exposure to the PEMS challenge. From 15 to 19 d of age, PCV in PEMS-exposed poults decreased to the level of the nonexposed control poults. At 11 d of age (5 d postexposure), the PEMSexposed poults receiving MC in their diet also showed a significant, but transitory, increase in PCV (39%; Figure  4 ) that was significantly less than the 45% peak in the PEMS-exposed poults not given MC. By 15 d of age, MC-treated PEMS-exposed poults showed a significant FIGURE 1. Weekly body weights for control, Myco Curb (MC), poult enteritis and mortality syndrome (PEMS), and Myco Curb + PEMS treatment groups. Unlike lowercase letters above the bars of the histograms indicate significant differences among the treatment groups (P < 0.05). decrease in PCV that returned to nonexposed control levels by 19 d of age (Figure 4 ).

Blood Hgb responses to PEMS and MC treatment (Figure 5) presented a profile similar to that of PCV ( Figure  4) . Nonexposed control and MC-treated poults showed Hgb levels between 9 and 10.5 g/dL throughout the trials. However, the blood Hgb levels of PEMS-exposed poults increased immediately by Day 7 (1 d after PEMS exposure) and peaked (14 g/dL) at 15 d of age. From 15 to 17 d of age, Hgb levels decreased significantly to levels comparable to those the nonexposed control and MCtreated groups but continued to show a significant decrease through 19 d of age ( Figure 5 ). The MC-treated PEMS-exposed poults experienced a small but significant increase and peak (11.5 g/dL) in blood Hgb levels 5 d (11 d of age) after exposure ( Figure 5 ). However, by 15 d of age the PEMS-exposed MC-treated poults had a lower Hgb level, and by 17 d of age the blood Hgb level had returned to nonexposed control and MC-treated levels. These levels continued to decrease through 19 d of age to a level that was significantly lower than nonexposed Hgb levels ( Figure 5 ). By 21 d of age, all treatment groups, with the exception of the PEMS only group, had similar blood Hgb levels ( Figure 5 ).

MC as a feed additive exerts its action primarily via propionic acid, a volatile fatty acid, when in its undissociated lipophilic form easily penetrates bacterial cell walls . Once inside the bacterial cell, the acid may dissociate and kill the cell . However, the effectiveness of propionic acid to control Enterobacteriaceae such as Salmonellae appears to depend on the severity of Salmonellae challenge (Hume et al., 1993) . Additionally, antibacterial properties of propionic acid diminish in the distal end of the intestinal tract as a result of reduced concentrations as it is absorbed, dissociated, and metabolized as a precursor in the tricarboxylic acid (TCA) cycle (Bolton and Dewar, 1964; Hinton et al., 1990; Hume et al., 1993) .

Anaerobic bacteria found in the intestinal flora of mature poultry produce volatile fatty acids as a natural mechanism of host resistance to Salmonellae infestation . However, young birds, especially during the first few days of life, do not have this established flora, and are more susceptible to pathogenic bacterial infections , such as those associated with PEMS.

Previous studies have shown that propionic acid probably influences appetite (Cave, 1978) and palatability (Cave, 1984) in chicks. However, when propionic acid is supplemented up to 3% in the diet, no significant effects on feed intake were noted (Fancher and Jensen, 1988) . Nevertheless, the data in this study indicated that the addition of 2.50% MC in diets given to young turkeys was not tolerated as well as 3.0% is in chicks. This observation was consistent with that made by Donaldson et al. (1994) who reported decreased feed intake and growth when 2 or 4% proprionate salts were given to newly hatched turkey poults.

Brake (Kemin Industries, 1993) indicated that the presence of a mold inhibitor may reduce microbial flora that are competitive with pathogenic bacteria. In one experiment, pathogenic bacteria caused a decrease in mortality in MC-treated (1% MC supplementation) broiler breeders during Weeks 12 to 14 compared to the control birds (Kemin Industries, 1993) . Thus, the same phenomenon might occur, which was indicated by the microbiological data from the cecal samples in the first experiment (Table  2) : MC at the 1.25 and 2.5% significantly reduced cecal populations of Enterobacteriaceae tribes and genera of bacteria. Therefore, a fine line exists between the quantity of feed additive required to reduce pathogenic bacteria and the quantity that possibly could affect commensal bacterial of the gut.

A systematic investigation to determine the effects of MC on PEMS-infected poults had not been conducted previously. MC fed at the 1.25% level in the diet of PEMSchallenged poults did not prevent the disease from occurring in the PEMS-challenged poults or the growth depression associated with PEMS. However, it did reduce the cumulative mortality by 50% and produced a delay in the onset of the initial mortality spike (Figures 2 and  3) . Dietary MC at 1.25% decreases bacterial load in the intestine and ceca. Therefore, it is possible that, at 1.25%, there was a decrease in the pathogenic bacterial load in the intestines or decreased colonization by various Enterobacteriaceae in PEMS-infected poults; however, the microbiological influence of MC in PEMS-infected poults was not determined in this study. Nevertheless, there are viruses such as newly identified and characterized turkey Astroviruses Shultz-Cherry et al., 2000 and 2001; Yu et al., 2000a,b) that induce a condition comparable to PEMS. It is, perhaps, possible that MC did not have any inhibitory influence on novel viruses associated with PEMS Shultz-Cherry et al., 2000 and 2001; Yu et al., 2000a,b; Heggen-Peay et al., 2002) , but its antibacterial influence possibly reduced the severity of enteric bacterial loads in PEMSchallenged poults in this study. The reduced bacterial load in control poults (Experiment 1) attributed to MC possibly is reflected in the overall decrease in mortality in MC-treated, PEMS-challenged poults (Experiment 2). However, the severity of the PEMS challenge in this study (1 mL of a 10% suspension of fecal matter from PEMS infected poults) was 10-fold greater than the PEMS challenge given in subsequent studies (0.1 mL of a 10% suspension of fecal matter from PEMS-infected poults) and might have been too large to be effectively suppressed by MC at 1.25%. Additional work with MC and its influence in PEMS appears to be warranted. Hemoconcentration, as indicated by a PCV of 45% in PEMS-infected poults was maximized 7 to 8 d after exposure. This peak coincided with the period of greatest diarrheal flushing, lack of appetite, dehydration, weight loss, and mortality similar to the observations made by Edens and Doerfler (1998) . All of these signs of PEMS disease contribute to reduction in blood volume due to severe dehydration leading to hemoconcentration and increased PCV. At 17 d of age, a more normal PCV developed in PEMS-infected poults. We hypothesized that this was a compensatory response in surviving poults as they attempted to rehydrate themselves.

The results from this study suggested that MC had a positive influence in PEMS-infected poults. The protective effect of MC extended beyond its influence on gut microflora and potential mycotoxins in feeds and to the reduction of the severity of PEMS-related bacterial-induced, but not novel viral-induced, diarrhea that contributed to the hemoconcentration associated with PEMS.

MC may have the potential to be more effective in the field, where poults are subjected to repeated challenges through bacterial contamination of feed, litter, or drinking water. In those common but hostile environments, where there was increased risk of exposure to overwhelming numbers of potentially pathogenic bacteria, the continuous presence of MC at low levels in the diets of poults at risk for development of PEMS may be sufficient to arrest the development of severe, if not lethal, bacterial enteritis.

Thus, MC might have some beneficial effect in improving the performance of turkey poults at risk for PEMS. Its place in the package of managerial procedures to control PEMS appears to be self-evident even though it was not the panacea that all turkey growers would be seeking. Nevertheless, MC has the ability to reduce significantly the early mortality due to PEMS and affects BW and feed conversions as well. With the growing concern about the prophylactic use of antibiotics, low level incorporation of MC into the feed of turkeys and other species of poultry appears to be a feasible low cost intervention against problems associated with bacterial enteritis.

Pages 1-11 in Poult enteritis and mortality syndrome (spiking mortality in turkeys) and related disorders: An update. College of Veterinary Medicine

The digestibility of acetic, propionic and butyric acids by the fowl

The influence of non-esterified fatty acids on feeding activity of chicks

Effect of dietary propionic and lactic acids on feed intake by chicks

Effect of dietary lactose on Salmonella colonization of market-age broiler chickens

Administration of propionate to day-old turkeys

Influence of Myco Curb and Sal Curb on feed intake and livability of turkey poults. Interim Report to the PEMS Task Force

Principles of ex ovo competitive exclusion and in ovo administration of Lactobacillus reuteri

Atypical Escherichia coli strains and their association with poult enteritis and mortality syndrome

Characterization of two Escherichia coli isolates associated with poult enteritis and mortality syndrome

Involvement of atypical Escherichia coli strains in turkey health

Poult enteritis and mortality syndrome: Definition and nutritional interventions

Effect of litter moisture and brooding temperature on growth and livability of turkey poults experiencing poult enteritis and mortality syndrome

Identification of Enterobacteriaceae

Induction of voluntary feed intake restriction in broiler chicks by dietary glycolic acid supplementation

High mortality and growth depression experimentally produced in young turkeys by dual infection with enteropathogenic Escherichia coli and turkey coronavirus

Alterations in macrophage-produced cytokines and nitrite associated with poult enteritis and mortality syndrome

Isolation of a PEMS-associated reovirus and its pathogenicity in turkey poults

Biological control of Salmonella typhimurium in young chickens

The effect of MycoCurb, propionic acid, and calcium propionate on the intestinal strength of broiler chickens

Metabolism of [ 14 C] propionic acid in broiler chicks

Inhibitory Effects of MycoCurb on various bacterial species. Publication No. 02326. Kemin Industries

Microorganism inhibition assays on Kemin MycoCurb. Publication No. 02332

Research Note: The effect of organic acids and Salmonella contamination originating from mouse fecal pellets

The post hatch physiology of turkey poults

Use of most probable number method modified with a deoxyribonucleic acid probe to monitor control by food preservatives of natural Salmonella contamination in animal meat meals

Immune system dysfunction during exposure to poult enteritis and mortality syndrome agents

A novel "small round virus" inducing poult enteritis and mortality syndrome and associated immune alterations

A PEMS-associated turkey astrovirus induces functional defects in macrophages

SAS/STAT8 User's Guide: Statistics

Modifying agent(s) associated with poult enteritis mortality syndrome: Importance of the thymus

Inactivation of an astrovirus associated with poult enteritis mortality syndome

Viral agents associated with poult enteritis and mortality syndrome: Role of a small round virus and turkey coronavirus

Characterization of a small round virus associated with the poult enteritis and mortality syndrome