Bovine respiratory disease complex (BRDC)
BRDC manifests from multiple causes and predisposing factors (Taylor et al., Reference Taylor, Fulton, Lehenbauer, Step and Confer2010; Caswell, Reference Caswell2014). Viruses, including bovine coronavirus (BCV), bovine herpesvirus 1 (BHV-1), bovine parainfluenza virus type 3 (BPIV-3), bovine respiratory syncytial virus (BRSV), and bovine viral diarrhea virus (BVDV) cause or predispose cattle to BRDC (Caswell, Reference Caswell2014). Bacteria, including Bibersteinia trehalosi, Histophilus somni, Mannheimia haemolytica, Mycoplasma bovis, and Pasteurella multocida, many of which are normal flora of the bovine upper respiratory tract, cause BRDC (Caswell, Reference Caswell2014). Additionally, a myriad of environmental factors contribute to BRDC, including transportation, sale barn auctions, and comingling (Cusak et al., Reference Cusak, McMeniman and Lean2003; Taylor et al., Reference Taylor, Fulton, Lehenbauer, Step and Confer2010). Thus, BRDC manifests from complicated interactions between and within individuals of a population, the stressors they are subjected to, and the microbes that they harbor, including those that act as opportunistic pathogens.
Mannheimia haemolytica
M. haemolytica is a primary agent of fibrinonecrotic pneumonia in cattle and a major bacterial component of BRDC (Rice et al., Reference Rice, Carrasco-Medina, Hodgins and Shewen2008); the bacteria is classified within the Pasteurellaceae family and is represented by distinct Gram-negative, nonmotile bacteria that may be either rods or coccobacilli, and that share a number of metabolic and/or biochemical phenotypes (Angen et al., Reference Angen, Mutters, Caugant, Olsen and Bisgaard1999a, Mutters et al., Reference Mutters, Angen, Bisgard and Garrity2005). When grown on blood agar plates, M. haemolytica colonies may be surrounded by a distinct zone of β-hemolysis, due to production of leukotoxin (LKT), which is a major virulence factor (Mutters et al., Reference Mutters, Angen, Bisgard and Garrity2005, Singh et al., Reference Singh, Ritchey and Confer2011). Importantly, M. haemolytica is a commensal organism of the bovine nasopharynx and tonsilar crypts (Shoo et al., Reference Shoo, Wiseman, Allan, Dalgleish, Gibbs, Al-Hendi and Selman1990; Radostits et al., Reference Radostits, Gay, Blood, Hinchcliff and Saunders2000). However, when animals are environmentally stressed and/or immuncompromized by viral infections, M. haemolytica can invade the lungs, evade the host immune response, and cause pulmonary inflammation with leukocyte damage and apoptosis due to LKT, lipopolysaccharide, and other virulence factors, and eventual death of the bovine host (Radostits et al., Reference Radostits, Gay, Blood, Hinchcliff and Saunders2000).
Considerable evidence shows that M. haemolytica bacteria are genetically and phenotypically diverse, and that distinct subsets of M. haemolytica associate with disease, whereas others do not (Frank and Smith, Reference Frank and Smith1983; Quirie et al., Reference Quirie, Donachie and Gilmour1986; Rice et al., Reference Rice, Carrasco-Medina, Hodgins and Shewen2008; Klima et al., Reference Klima, Alexander, Hendrick and Mcallister2014). M. haemolytica are typically encased in capsules, which are virulence factors that interfere with host cellular phagocytosis (Chae et al., Reference Chae, Gentry, Confer and Anderson1990). There are 12 capsular serotypes for M. haemolytica (A1, A2, A5, A6, A7, A8, A9, A12, A13, A14, A16, and A17) (Angen et al., Reference Angen, Quirie, Donachie and Bisgaard1999b). Isolates of serotypes A2 and A4 are commonly found in the upper respiratory tract of healthy cattle, and rarely in the lungs of diseased animals. Conversely, isolates of serotype A1 in particular, and also those of serotype A6, are predominantly overrepresented in the lungs of diseased animals. See Rice et al. (Reference Rice, Carrasco-Medina, Hodgins and Shewen2008) and Singh et al. (Reference Singh, Ritchey and Confer2011) for reviews of serotype associations.
In addition to capsular serotyping, M. haemolytica diversity has been characterized using pulsed-field gel electrophoresis (PFGE) (Klima et al., Reference Klima, Alexander, Read, Gow, Booker, Hannon, Sheedy, Mcallister and Selinger2011, Reference Klima, Alexander, Hendrick and Mcallister2014; Timsit et al., Reference Timsit, Christensen, Bareille, Seegers, Bisgaard and Assié2013). This is a technique that detects insertions and/or deletions, genome rearrangements, and, to a smaller extent, nucleotide polymorphisms within microbial genomes through restriction enzyme digests and electrophoretic banding profiles (Kudva et al., Reference Kudva, Evans, Perna, Barrett, Ausubel, Blattner and Calderwood2002; Foley et al., Reference Foley, Lynne and Nayak2009; Goering, Reference Goering2010; Sabat et al., Reference Sabat, Budimir, Nashev, Sá-Leão, Van Dijl, Laurent, Grundmann and Friedrich2013). M. haemolytica isolates originating from the lungs of diseased animals have been found to have PFGE banding profiles that suggest closer genetic similarity to one another than to isolates originating from the upper respiratory tract of non-diseased animals (Klima et al., Reference Klima, Alexander, Hendrick and Mcallister2014). Consequently, two very different techniques that assess microbial diversity, capsular serotyping, and PFGE, both support the notion that M. haemolytica subtypes do not share an equal propensity for associating with, or causing, respiratory disease in cattle. This also indicates that a potential approach to managing or preventing bovine respiratory disease attributable to M. haemolytica could involve testing cattle for the subtypes they are harboring, and isolating and/or treating those that are harboring strains with an increased propensity to cause disease.
Goals and approach
The goals of this ongoing project are to (1) identify fundamental genetic variation within M. haemolytica strains of North America; (2) develop a set of nucleotide polymorphisms to detect that variation; and (3) use the set to identify genetic determinants that influence M. haemolytica pathogenicity in cattle. This is being accomplished through whole genome sequencing of M. haemolytica strains that originated from either the upper or lower respiratory tract of cattle that were either (1) clinically ill or deceased, or (2) disease free. Two collections of M. haemolytica strains have been assembled that are comprised of strains isolated from the lungs of clinically ill or deceased cattle throughout North America (Portis et al., Reference Portis, Lindeman, Johansen and Stoltman2012). One collection is being used primarily for the discovery of genetic variation, including nucleotide polymorphisms, and consists of 158 epidemiologically unlinked strains originating from cattle within 35 US states, as well as the Canadian provinces of Alberta, British Columbia, Manitoba, Ontario, and Saskatchewan. The strains were isolated from beef or dairy cattle from 2002 to 2011, with a majority originating from beef animals. The second collection is being used primarily for validation of the discovery collection, and consists of 163 epidemiologically unlinked strains originating from 29 US states, as well as the Canadian provinces of Alberta, Ontario, and Saskatchewan. Strains of the validation collection were isolated from beef or dairy cattle from 2002 to 2011, with a higher representation of dairy cattle germplasm than the discovery population. These two collections represent a deep sampling of North American M. haemolytica genetic subtypes that cause or associate with respiratory disease. Additionally, a third M. haemolytica collection has just recently been assembled that is comprised of strains isolated from the nasopharnynx of US cattle that were not afflicted with respiratory disease.
Whole genome sequencing of the M. haemolytica strains from the collections described above will reveal genetic diversity across the entire genome that will serve as a roadmap for defining M. haemolytica subtypes, and for testing them for an association with respiratory disease. Tests will be developed that will identify M. haemolytica subtypes. The results from this study will be published and placed into the public domain for use without restriction.
Acknowledgments
The authors thank diagnostic labs at the following institutions for their donations of M. haemolytica strains: Cornell University, Iowa State University, Kansas State University, Michigan State University, North Carolina Veterinary Diagnostic Center, Ohio Animal Disease Diagnostic Laboratory, Oklahoma State University, Pennsylvania State University, South Dakota State University, Texas A&M University, University of Saskatchewan, UC Davis, University of Guelph, University of Illinois, University of Minnesota, the University of Wisconsin, and Washington State University. The authors also thank Professor J. Dustin Loy and the University of Nebraska for donated strains.
