Mycoplasmas are the smallest free-living bacteria and lack a cell wall (Razin, 1996). Mycoplasma bovis (M. bovis) is the most pathogenic mycoplasma in cattle (Caswell et al., 2010). It is well adapted to live on mucosal surfaces such as the respiratory tract and mammary glands (Peek et al., 2018). M. bovis can cause pneumonia, mastitis, mid-ear infection, arthritis, and other diseases. Cattle are most vulnerable during stressful conditions such as lactation and transport (Jasper et al., 1974; Arcangioli et al., 2008). The estimated losses caused by M. bovis from involuntary culling, reduced feed conversion and decreased production, medical treatments, extra labor, etc. are more than $2 billion/year worldwide (Maunsell et al., 2011). Outbreaks of M. bovis diseases have been reported in virtually every country, the most recent being New Zealand (July 21, 2017), a country with no prior records of M. bovis diseases according to their Ministry of Primary Industries (Boyce et al., 2021).

M. bovis is resistant to commonly used antibiotics because it lacks a cell wall, does not synthesize nucleotides or amino acids (Wormser and Hardy, 2005), and has an unusual form of RNA polymerase (Razin et al., 1998). Although M. bovis is susceptible to a group of uncommon and expensive antibiotics such as enrofloxacin (Rosenbusch et al., 2005), resistance is increasingly observed (Lysnyansky and Ayling, 2016). For example, Cai et al. (2019) found that the minimum inhibitory concentrations (MIC₅₀) of chlortetracycline, oxytetracycline, tilmicosin, and tylosin were much higher in 2009 than the values found in 1990 demonstrating decreased susceptibility of M. bovis to antibiotics over time. Even resistance to enrofloxacin was also reported (Sulyok et al., 2017).

While several vaccines have been developed, most showed inconsistent effectiveness (Nicholas et al., 2002; Perez-Casal et al., 2017), which is attributed to the fast recombination of surface lipoproteins (Caswell and Archambault, 2007). The first commercial vaccines analyzed were Mycomune^® R (BIOMUNE Co., Lenexa, KS, United States) and Pulmo-Guard^™ MpB (American Animal Health, Inc.). Both were shown to be ineffective against M. bovis colonization of the upper respiratory tracts or development of lesions caused by M. bovis. Their efficacies were shown to be 44% and less than 1%, respectively (Soehnlen et al., 2011). Furthermore, the two currently available vaccines, MpBGaurd and Myco-B One Dose (Dudek et al., 2021), contain two and three field isolates and their soluble antigens, respectively. Myco-B One Dose was claimed to prevent death and morbidity by only 10.4 and 15.2%, respectively.¹ With the lack of effective vaccines (Gautier-Bouchardon et al., 2014; Yair et al., 2020), the use of alternative prevention and treatment regimens for M. bovis is therefore called for.

Plant-derived antimicrobials (PDAs) such as trans-cinnamaldehyde (TC), carvacrol (CAR), and eugenol (EU) are generally considered safe (GRAS) to humans/animals yet possess strong anti-bacterial and anti-fungal effects (Michiels et al., 2007). For instance, cinnamaldehyde and EU have antimicrobial effects in a wide range of bacteria that infect cattle such as Mannheimia haemolytica, Escherichia coli, Staphylococcus aureus, and M. bovis (Rajamanickam et al., 2019). All three PDAs have also been shown to be inhibitory on numerous species of fungi such as Cryptococcus neoformans, Cryptococcus laurentii, and Microsporum gypseum (Kumari et al., 2017; Michalczyk and Ostrowska, 2021).

Although not completely understood, the mechanisms of PDAs actions may include breaking down membrane integrity/increasing permeability, disrupting energy production, and other metabolic regulatory functions (Swamy et al., 2016). Whether these compounds modulate gene expressions of mycoplasma, however, has not been reported.

In preliminary studies, we found that, at low doses, TC (0.02%), EU (0.08%), and CAR (0.04%) caused growth inhibition of M. bovis in culture. This study was designed to delineate the molecular mechanisms for such inhibition by transcriptomic analysis. We found CAR exhibited ample effects on gene expression, while TC and EU might exert growth inhibition through other mechanisms than gene expression regulations.

To our knowledge, this is the first transcriptomic study on M. bovis. While a previous study showed that EU and TC exhibited growth inhibition of M. bovis in vitro (Rajamanickam et al., 2019) which agreed with our preliminary growth inhibition results, the molecular mechanism of the PDA-mediated inhibition was not determined. Here, we showed that CAR’s inhibitory action on M. bovis involved differential expression of a large number of genes.

All PDAs used in our study have been linked with damage to cell membrane integrity, polarity, and free-radical generation in microbes (Devi et al., 2013; Shen et al., 2015; Al Wafai et al., 2017; Khan et al., 2017). It is therefore possible that membrane disruption may be the first action of PDAs at levels used in causing cell growth inhibition or cell death. Other actions such as altering gene expression regulation may be secondary. This could be the reason that we did not find large-scale gene expression changes for most PDA treatments (i.e., no DEGs from EU, TC + CAR, and Triple treatments, and very few from EU + CAR, TC + EU, and TC treatments).

A number of DEGs in CAR-treated samples are notable given their annotated function. A BspA family leucine-rich repeat surface protein is among the downregulated genes. BspA has been associated with Tannerella forsythia bacterial adhesion and invasion of epithelial cells in vitro (Inagaki et al., 2006). Downregulation of BspA upon CAR treatment weakens M. bovis’ pathogenicity and may be a mechanism of CAR’s antimicrobial efficacy. Downregulation of recU, which plays an important role in chromosome segregation and DNA damage repair (Pereira et al., 2013), may be another potential mechanism for CAR’s antimicrobial action.

Nucleotidyltransferases are a family of enzymes that add nucleotides to substrates like nucleic acids and proteins (Neuenfeldt et al., 2008), and have involvement in DNA repairs (Liu et al., 2007). Upon CAR treatment nucleotidyltransferase is downregulated. This agrees with previous work which showed that the stress of 1,000 mg/l phenol decreased the expression of nucleotidyltransferase in Acinetobacter strain Y (Lin, 2017). Because CAR is a phenolic compound (Sharifi-Rad et al., 2018), this downregulation of nucleotidyltransferase is therefore a potentially logical mechanism for CAR’s antimicrobial effect by impairing DNA repair/synthesis. Another potential effect of CAR on the pathogenicity of M. bovis may be the downregulation of variable surface lipoproteins (Supplementary Table 2), which are assumed to be associated with the capacity of bacteria to adhere to epithelial cells and evade immune system (Christodoulides et al., 2018). This potential effect of CAR, although not related to growth inhibition, could significantly contribute to its overall antimicrobial effect (Michiels et al., 2007). Notably, combined EU + CAR treatment upregulated variable surface lipoprotein expression, but in genes paralogous to that affected by CAR treatment alone.

Phosphotransferase system (PTS) transporter subunit EIIB is part of the bacterial PTS system involved in uptake of sugar molecules like fructose, mannitol, and cellobiose (Deutscher et al., 2014). The downregulation of PTS by CAR treatment indicates disruption of M. bovis sugar uptake, which potentially reduces glycolysis and ATP generation. This is further exacerbated by the actual downregulation of ATP-binding proteins found in our GO analysis (Figures 2A,C). Gene prs is involved in synthesis of the intermediates for purine and pyrimidine, nicotinamide adenine dinucleotide synthesis (Hove-Jensen, 1988), whereas ftsH3 involves in degradation of various transmembrane and cytoplasmic proteins (Srinivasan et al., 2008). Similarly, cytidylate kinase is one of the enzymes that are involved in nucleotide metabolism (An et al., 2014). This shows treatment with CAR results in disruption of cellular metabolism, including synthesis of DNA and RNA. Interestingly, when M. bovis was treated with EU + CAR, PTS transporter subunit EIIB expression was upregulated; this may be due to an antagonistic effect of CAR by EU (Bassolé and Juliani, 2012).

Moreover, we found CAR to affect expression of genes in the pentose phosphate pathway, including deoC and prs which were both downregulated by CAR. deoC is known to catalyze the reaction of 2-deoxy-D-ribose-5-phosphate into glyceraldehyde-3-phosphate and acetaldehyde (Haridas et al., 2018), while prs is the enzyme that generates phosphoribosyl diphosphate (Hove-Jensen et al., 2016). The inhibition of these enzymes conceivably reduces DNA replication and RNA synthesis, thus decreasing growth and replication of M. bovis in response to CAR.

Data here indicated that a major effect of CAR on M. bovis is to alter the translation process. However, CAR’s effect on translation-related genes was not uniform because different genes in the process could be either up- or downregulated. For example, serine- tRNA ligase and rRNA pseudouridine synthetase were downregulated by CAR, yet six ribosomal protein genes (rplF, rpsR, rplK, rplA, rpsR, and rplL; Kaczanowska and Rydén-Aulin, 2007) were upregulated, as was the pre-tRNA and rRNA processing enzyme rnc (Li, 2013). Similar conflicting effects of both up- and downregulation of different genes in the same pathways such as ATP generation (Guo et al., 2019; upregulation of ATP synthase subunits F0F1 ATP synthase subunit A and B) and pentose phosphate pathway (Lyngstadaas et al., 1998; upregulation of the ribulose-phosphate 3-epimerase) also occurred. These conflicting responses of M. bovis upon CAR treatment suggest that cells may have undergone chaotic changes of cellular processes and attempted to counter the reduction in growth by increasing protein translation and other inhibited processes.

Finally, we observed differential expression of many hypothetical proteins in response to CAR. Given the novelty of these proteins and their differential regulation in response to an inhibitory substance like CAR, they may be of great importance regarding survival or pathogenicity of M. bovis and represent important targets for direct experimentation in laboratory (Galperin, 2001). The identification of the functions of these hypothetical proteins will further our understanding of the effects of CAR.

In conclusion, at growth-inhibitory levels, TC, EU, and CAR exhibited different effects on gene expression by M. bovis with CAR inducing the most changes. It is likely that TC and EU at levels studied exerted their inhibitory effects through other mechanisms than that exerted by CAR. The antimicrobial effects of CAR involved the reduction of (1) certain translation-related enzymes, (2) DNA/RNA synthesis through pentose phosphorylation pathway, (3) surface proteins and lipoproteins associated with bacterial adhesion to and invasion of host tissues, and (4) ATP generation.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at: Gene Expression Omnibus (GEO)—GSE198086.

SR, JD, KS, IM, and FA conducted the experiments and/or data analysis. KV, ET, and XT designed the study. XT obtained research funding. All authors contributed to the drafting/editing of the manuscript. All authors contributed to the article and approved the submitted version.

This study was supported by USDA grants to XT (58-8042-5-047 and W4171) and NSF grant to IM (1564936).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.