Determination of tylosin concentration in sow’s milk after intramuscular administration

Initially isolated in 1961, tylosin (Chemical Abstracts System number 1401-69-0) is a traditional representative of the class of macrolides, a significant group of antibiotics secreted by soil-borne bacteria belonging to the Streptomyces genus (29). Structurally, the antibiotic consists of a typical macrolactone ring containing 16 carbon atoms (a tylonolide), with mono- and disaccharide branches attached (18). Its trade name has traditionally been used for a mixture of four pharmacologically active derivatives produced by S. fradiae through fermentation: tylosin A, the major component of the mixture and most commonly used marker residue; and three minor factors – tylosin B (desmycosin), tylosin C (macrocin), and tylosin D (relomycin) (15, 21). The relative bioactivities of the constituents are 1, 0.83, 0.75 and 0.35, respectively (34).

Similarly to the majority of members belonging to the macrolide class, the antimicrobial mode of action of tylosin is exerted by its reversible binding to the 23S rRNA component of the 50S ribosomal subunit, thus effectively preventing protein synthesis in susceptible bacteria (3). An alkaline environment was found to increase its antimicrobial activity. Tylosin’s pharmacokinetics has been defined in different species of animals (16, 17, 25) and characterised in the available literature by a high volume of distribution throughout tissues and body fluids, after both oral and parenteral administration of the drug. Tylosin binds to plasma proteins at a slight to moderate level. Its half-life in most animal species was defined as 3 to 4 h, and its excretion is mostly via faeces and bile (29).

Tylosin presents antimicrobial properties typical of the class. Active against mycoplasma, most Gram-positive bacteria (such as Streptococcus and Staphylococcus species), and several Gram-negative bacteria, it is described as a medium-spectrum antibiotic. Tylosin is predominantly considered a bacteriostatic antimicrobial agent; however, it was found to have bactericidal properties at elevated concentrations. Therapeutic usage of tylosin is strictly limited to veterinary medicine. Commercially important, the antibiotic has been authorised for marketing in several countries in the forms of tylosin base and its phosphate and tartrate salts, administered via intramuscular injection or orally, in the drinking water or after incorporation into medicated feed (29).

Early studies performed on several in-feed antibiotics including tylosin given to commercially reared swine reported significant improvements in both feed efficiency and growth performance of the animals (5). Nevertheless, owing to the well-documented environmental pollution and substantial risk of the spread of antibiotic resistance (14, 22), the usage of tylosin at subtherapeutic levels as a feed additive for growth promotion in food-producing animals was eventually banned in the EU in Council Regulation (EC) No. 2821/98 of 17 December 1998.

Today, the antibiotic is available under a broad range of commercial names, and has found clinical applications in the treatment of various bacterial conditions, including mycoplasmal diseases in major food-producing animals (13, 35). Tylosin is also among the antibiotics most widely used to treat infections caused by the Gram-negative anaerobes Lawsonia intracellularis (23) and Brachyspira hyodysenteriae (11), respectively the causative agents of porcine proliferative enteropathy and swine dysentery. However, research performed on field isolates from swine of the latter bacterium have demonstrated dwindling efficacy of tylosin (24, 36). Besides the antibiotic having found therapeutic applications in less obvious food-producing animals, it is used to control Paenibacillus larvae, the bacterium causing American foulbrood in the honey bee (31).

Regardless of the establishment of multiple analytical protocols allowing an affordable, accurate and rapid determination of tylosin (both exclusively and in a multianalyte method) in a wide variety of biological matrices including meat and offal (19, 27), dairy products (27), eggs (27), aquatic products (4), honey (19), feed (30), manure (12) and porcine oral fluid (9, 28), research evaluating the antibiotic concentration in sow’s milk cannot be found in scientific literature. Hence, the purpose of this study was to fill in the knowledge gaps and determine the concentration–time profile of tylosin in this matrix.

Pharmacokinetics data on the transfer of tylosin into milk are restricted to several peer-reviewed works carried out using cows and less obvious dairy animals as study models (1, 8). It is relevant that apart from differing in the ruminant species included, the research frameworks in the experimental studies consisted of different administration routes, antibiotic dosages and sampling intervals. Moreover, as a natural consequence of practical focus on the treatment of the udder tissue infections in ruminants, the choice of animals involved in clinical trials was most often confined to those affected by mastitis. The lack of tabular data on actual residue concentrations of the antibiotic observed during a study and the adoption of different analytical procedures (including less accurate methods such as microbiological assays) in any such studies also severely limit direct comparisons. Even though the accumulated evidence seems to be radically influenced by all the aforementioned aspects, the data obtained in our study concurred well with previous findings demonstrating extensive penetration of tylosin to milk in ruminants.

As a weak lipophilic organic base, tylosin easily passes the blood–milk barrier, reaching its maximum concentration in a relatively short time. Indeed, as seen from our investigation, the highest mean concentration of the antibiotic among all the samples (1,802 μg/L; SD 349) was attained at 3 h following the administration of the therapy (Table 2). In a study involving clinically healthy goats which had received a single intramuscular dose of 10 mg of tylosin per kg of live weight, the antibiotic was detectable from 2 to 10 h after administration, at levels sharply declining along the timeline from 2,940 μg/L to 160 μg/L (1). Moreover, the concentration reported at 4 h (1,730 μg/L) closely matched the value obtained in our study at 3 h (1,802 μg/L). In another piece of research conducted on clinically healthy Slovak merino ewes injected with the same dose of the antibiotic for five consecutive days and milked for samples every 12 h, the highest mean concentration of tylosin (1,822 μg/L) was determined 12 h after the second administration (26).

In the present study, a rapid distribution of the antibiotic to sow’s milk was followed by its marked decline in this matrix to the levels of 744 μg/L (SD 371) on day 1 (24 h) and 482 μg/L (SD 140) on day 2 (48 h). Tylosin was also demonstrated to still be at measurable residual levels in samples collected on day 5 (120 h) and day 7 (168 h), but no residue of the antibiotic was found in the milk of any sow on day 14 (336 h) or day 21 (504 h). In other words, the last sampling time point when the residues were detected was 120 h after the last treatment. In previously cited research, the last sampling intervals at which tylosin could still be detected were 36 and 12 h after termination of the therapy in ewes and goats, respectively (1, 26). Nevertheless, research demonstrating a significantly longer decline time (96 h) in the latter species given a single dose of 10 mg of tylosin per kg is also on record (32).

It has been noted that in general, tylosin residues are depleted faster in small ruminants than they are in cows. Despite the dissimilarity in the doses of tylosin being given via intramuscular administration to heathy lactating cows in two pieces of research, a typical excretion time for the antibiotic evaluated in their milk could be defined and ranged from 24 to 96 h (2, 10). According to publicly available literature, local alterations of physicochemical parameters in the affected mammary gland (defined by a somatic cell count) were found to have influenced the pharmacokinetics of the antibiotic; cows which had contracted different types of mastitis exhibited significantly extended excretion periods, retaining measurable residual levels of tylosin in their milk up to 10 days (240 h) after termination of the therapy (8, 20).

As noted above, tylosin penetrated the udders of lactating sows well and relatively quickly when the pigs were given an intramuscular injection of the antibiotic. Modern scientific knowledge about the duration and magnitude of antibiotic milk levels in sows remains fairly rudimentary and is represented exclusively by some few peer-reviewed studies published in recent years (6, 7); therefore, our research contributes to the canon of knowledge of the comparative pharmacokinetics of the antibiotic. Nevertheless, a systematic analysis of the factors which may have influenced the distribution of the drug in lactating sows and its elimination from them merits further engagement.

Tylosin residues in different animal tissues have been extensively studied to date. Although the maximum residue limit (MRL) established by the EU authorities for muscle, fat, liver and kidney tissue (100 μg/kg; Commission Regulation No. 37/2010) is lower than the one set by the USA (200 μg/kg; Code of Federal Regulations Title 21 Part 556), the same criterion of 50 μg/kg for milk was adopted in both regions. As seen in our study, values exceeding the norm were observed from the first day of the investigation. Mean concentrations of the antibiotic in milk samples at or exceeding the MRL established for tylosin in cattle milk were noted between day 5 (120 h) and d 7 (168 h) after the commencement of the therapy. Given the human food safety assurance motivation for which the norms were set, extrapolation of the standards to suckling piglets ingesting contaminated sow’s milk is of debatable validity.

In the view of the ineffectiveness of tylosin against Enterobacteriaceae and the (erroneous) belief that sow’s milk concentrations of antibiotics in general were quantitatively negligible, the lack of previously published research on the discussed issue seems to be justified. It is noteworthy, however, that selected members of the macrolide group have been found to produce a variety of sub-inhibitory effects – meaning effects when used at weaker strengths than the minimum inhibitory concentration (MIC) – on selected bacteria, including one considerably hindering bacterial motility and adherence (33). Nevertheless, the available research recognised the biological effects of individual macrolides rather than the whole class, reported effects only with certain microorganisms, and gained them using highly specific experimental conditions and mostly in vitro models.

The key findings and conclusions drawn in the literature cannot be extrapolated either to swine intestinal pathogens or to tylosin itself, because that research was not analogous to the present investigation. This being the case, the potential role of sub-MICs of tylosin in suckling piglets, including post-exposure effects on the development of intestinal microbiota, deserves further examination.