Effects of treatment with enrofloxacin or tulathromycin on fecal microbiota composition and function of dairy calves

The increasing concerns with antimicrobial resistance highlights the need for studies evaluating the impacts of antimicrobial use in livestock on antimicrobial resistance using new sequencing technologies. Through shotgun sequencing, we investigated the changes in the fecal microbiome composition and function, with a focus on functions related to antimicrobial resistance, of dairy calves. Heifers 2 to 3 weeks old, which were not treated with antibiotics by the farm before enrollment, were randomly allocated to one of three study groups: control (no treatment), enrofloxacin, or tulathromycin. Fecal samples were collected at days 4, 14, 56 and 112 days after enrollment, and DNA extraction and sample preparation and sequencing was conducted. The effect of antibiotic treatment on each taxon and functional level by time (including Day 0 as a covariate) revealed few changes in the microbiota. At the genus level, enrofloxacin group had higher abundance of Blautia, Coprococcus and Desulfovibrio and lower abundance of Bacteroides when compared to other treatment groups. The SEED database was used for functional analyses, which showed that calves in the enrofloxacin group started with a higher abundance of “Resistance to antibiotics and toxic compounds” function on Day 0, however an increase in antibiotic resistance genes after treatment with enrofloxacin was not observed. “Resistance to Fluoroquinolones” and “Erythromycin resistance”, of relevance given the study groups, were not statistically different in abundance between treatment groups. “Resistance to fluoroquinolones” increased during the study period regardless of treatment group. Despite small differences over the first weeks between treatment groups, at Day 112 the microbiota composition and functional profile was similar among all study groups. These findings show that metaphylaxis treatment of dairy calves with either enrofloxacin or tulathromycin have minimal impacts on the microbial composition and functional microbiota of dairy calves over time.


24
The effect of antibiotic treatment on each taxon and functional level by time (including Day 0 as 25 a covariate) revealed few changes in the microbiota. At the genus level, enrofloxacin group had 26 higher abundance of Blautia, Coprococcus and Desulfovibrio and lower abundance of Bacteroides 27 when compared to other treatment groups. The SEED database was used for functional analyses, 28 which showed that calves in the enrofloxacin group started with a higher abundance of "Resistance 29 to antibiotics and toxic compounds" function on Day 0, however an increase in antibiotic resistance 30 genes after treatment with enrofloxacin was not observed. "Resistance to Fluoroquinolones" and 31 "Erythromycin resistance", of relevance given the study groups, were not statistically different in 32 abundance between treatment groups. "Resistance to fluoroquinolones" increased during the study 33 period regardless of treatment group. Despite small differences over the first weeks between

43
There is urgent need for the judicious use of antimicrobial drugs to extend the effectiveness 44 of currently available therapies [1,2]. Antimicrobial resistance (AMR) is a natural phenomenon, 45 and resistance genes are present even in places that were never inhabited by humans [3,4] and in 46 wild animals [5]. However, the use of antibiotics in human and animal medicine, as well as in plant  Antimicrobials may be used to control and prevent the spread of the disease in production 54 animals at high risk of developing a bacterial infectious disease. This practice is referred to as 55 metaphylaxis. In the U.S., several drugs are approved for this use in cattle, including in dairy 56 calves, at risk of bovine respiratory disease (BRD). BRD is known to be a common cause of disease 57 in dairy calves at 2-3 weeks of age [9,10] . In 2014 in the United States 12% of pre-weaned calves 58 were affected with respiratory disease and almost 95% of those were treated with antimicrobial 59 drugs [11]. Macrolides and florfenicol were the drugs of choice on 18.2 and 15.1 percent of the 60 farms to treat BRD, followed by penicillin (8.1%) and fluoroquinolones (6.6 %) [11]. In our study, 61 we focused on enrofloxacin and tulathromycin, a fluoroquinolone and a macrolide, respectively.

62
These are injectable, single dose antimicrobials, labeled for treatment and control of BRD.

63
Fluoroquinolones are broad-spectrum antimicrobial drugs that target essential bacterial 64 enzymes DNA gyrase and DNA topoisomerase IV. Mutations in these bacterial enzymes can 65 confer resistance, as can plasmids carrying resistance genes that protect the cells from the 66 bactericidal effects of quinolones [12,13]. Fluoroquinolones were developed about 40 years ago fluoroquinolone, enrofloxacin, for dogs and cats [16]. Later, in 1998, it was approved for treating 71 bovine respiratory disease in cattle. The use of enrofloxacin in poultry was banned in 2005 and in 72 2008 its use in female dairy cattle was restricted to nonlactating animals up to 20 months old.

73
Extra-label use of enrofloxacin is strictly prohibited. Enrofloxacin has also been approved for 74 treatment and control of swine respiratory disease since 2008. [16]. 75 Macrolides are mainly bacteriostatic. They inhibit bacterial protein synthesis and the 76 spectrum of action of drugs within this class varies. In human medicine, erythromycin has been 77 available since 1952, and other current drugs in the same class are azithromycin, clarithromycin 78 and telithromycin [17]. In veterinary medicine, erythromycin and tylosin are used to treat 79 companion animals, ruminants, swine, and poultry [16]. Tulathromycin, the macrolide used in our 80 study, was approved by the FDA in 2005 to treat and control respiratory disease in cattle and swine 81 [16].

118
A randomized field trial study design was used. Calves 2 to 3 weeks old, which had not 119 been treated with antimicrobials before enrollment, were randomly allocated within cohort 120 (enrollment day) to one of three study groups: control (CON, no antimicrobial), enrofloxacin              At level 3, within "RATC", we found no significant difference in the relative abundance 302 of "Resistance to Fluoroquinolones" among treatment groups (P-value 0.19). Abundance of the 303 "RATC" function increased over time for all treatment groups, including for calves that did not 304 receive antibiotic treatment, and it was higher on Day 56 for calves treated with tulathromycin 305 (Figure 9). Also within "RATC", the relative abundance of "Multidrug efflux pump in 306 Campylobacter jejuni (CmeABC operon)" was significantly different among groups (P-value 307 0.015), with calves treated with enrofloxacin having higher abundance on Day 4 and then lower 308 on Day 56. There was no significant difference detected in "Erythromycin resistance" among 309 treatment groups (P-value 0.662). Our results support that metaphylactic treatment of preweaned calves with enrofloxacin or 336 tulathromycin did not cause major changes in fecal microbiota composition and gene functions 337 related to antimicrobial resistance. Despite the fact that calves were not treated with antibiotics 338 before enrollment and were randomly assigned to one of the three study groups, these groups 339 started the trial with different microbiota composition and function, including higher abundances 340 of "Virulence, Disease and Defense" and "RATC" functions in calves assigned to enrofloxacin 341 treatment. Nevertheless, after controlling for this difference in the analysis, we did not observe a 342 significant increase in the abundance of genes with these functions after treatment with 343 antimicrobial drugs.

344
When lactating cows are treated with antimicrobial or other drugs due to illnesses, their 345 milk is withheld from sale because it may contain drug residues. This non-saleable milk, also called 346 waste milk, is often fed to calves. Antibiotic residues found in waste milk from 34 farms in New 347 York state were mainly β-lactams, tetracycline and sulfamethazine [25]. Enrofloxacin is approved 348 only for non-lactating female dairy cattle less than 20 months of age, and extra-label use of this 349 drug is strictly prohibited; therefore, there should not be residues from this drug class in milk fed 350 to calves. Additionally, enrofloxacin was not used at the study farm as a treatment for calf 351 pneumonia before the study began. and Defense" and "RATC", as we hypothesized. On the other hand, calves treated with 360 enrofloxacin had higher abundance of "Transposable elements" on Day 4. Also known as 361 transposons, these mobile genes can transfer functions within genomes and cause mutations [32].

362
The enrofloxacin group also had higher abundances of "Clustering-based subsystems" and lower  which is both important for cell physiology and toxic to epithelial cells [44]. They are also 396 opportunistic pathogens found in the environment [45] and GI tract of humans and other animals 397 [46,47]. Because they are difficult to culture, there is a lack of information about this gram-18 398 negative anaerobe, including antimicrobial susceptibility data. In a study with 23 Desulfovibrio 399 isolates from four different species cultured from human specimens, D. fairfieldensis strains 400 showed resistance to β-lactams and the fluoroquinolones ciprofloxacin and levofloxacin [48]. 401 Culture-independent studies will bring more knowledge to these pathogens, but culture-based 402 methods continue to be important for antimicrobial susceptibility surveillance.

403
The genus Blautia has been associated with human gut health, being reduced in patients . These two genera were temporarily increased in calves that received 408 enrofloxacin in our study, which may be considered a good, unpredicted side effect.

409
Pitfalls of our study include the fact that, although calves were randomly enrolled in 410 treatment groups, initial microbial composition and function still differed between calves in 411 different treatment groups at Day 0. This study was performed on a single farm, which did not 412 allow comparisons of effects across herds. Nevertheless, the calf management practices of the farm 413 used in the study are typical for many other commercial dairy farms in New York State.

414
In our study, enrofloxacin or tulathromycin had minimal impacts on the microbial 415 composition and functional microbiota of calves over the study period (112 days) when used to 416 prevent and control BRD. It is important to note that "Resistance to fluoroquinolones" increased 417 during the study regardless of treatment group and, therefore, more efforts should be dedicated to 418 reduce the use of medically important antimicrobials in dairy calves.