Government License Rights

This invention was made with government support under contract number 2013-68004- 20362 awarded by United States Department of Agriculture. The government has certain rights in the invention.

BACKGROUND

Porcine reproductive and respiratory syndrome virus (PRRSV) and porcine circovirus type 2 (PCV2) are two of the most significant pathogens of swine worldwide. PRRSV, a single- stranded RNA virus in the family Arteriviridae , is widely considered to cause the most costly disease of swine in the United States (U.S.), with estimated annual losses of $664 million due to diminished weight gain and respiratory disease in growing pigs. PCV2, a single-stranded DNA virus in the family Circoviridae , is estimated to cause economic losses up to $20/pig in unvaccinated herds due to a group of syndromes termed porcine circovirus associated disease (PCVAD), which includes muscle wasting, weight loss and respiratory disease. Both PRRSV and PCV2 result in systemic infections and modulation of host immunity, reducing the rate of weight gain and increasing the likelihood of primary and secondary polymicrobial disease syndromes in swine.

PRRS modified live virus (MLV) vaccines are widely used in PRRS-endemic herds to reduce losses associated with PRRSV infection. In experimental and field settings, PRRS MLV immunization has the potential to improve weight gain, reduce viral replication, reduce pulmonary pathology, and decrease clinical disease after wild-type PRRSV exposure. However, several challenges remain for PRRS MLV vaccine safety and efficacy, including the potential for reversion to virulence and recombination with wild-type strains, potentiation of primary and secondary pathogens, and incomplete protection against emerging wild-type strains. As such, the currently available commercial vaccines are generally considered inadequate for disease control and improved vaccines or alternatives strategies for PRRS control are urgently needed.

The gut microbiome, or community of microorganisms in the gastrointestinal tract, is an emerging alternative tool for improving the response of swine to PRRS. Previous work has demonstrated several gut microbiome characteristics that are associated with several improved outcome parameters in pigs co-infected with PRRSV and PCV2. Specifically, increased gut microbiome diversity, increased Ruminococcaceae species, increased Streptococcaceae species, and fecal microbiota transplantation were associated with infection outcomes such as reduced virus replication, improved weight gain, and decreased morbidity. The mechanism by which the gut microbiome impacts outcome in pigs exposed to respiratory pathogens is largely unknown but is believed to be associated with modulation of immunity and through the metabolic products of microbes.

Considering the proposed mode of action for the microbiome on modulating immunity, the inventors reasoned it may be possible to extend the potential impact of gut microbes beyond primary infection, and to weight gain prior to or following vaccination. Research focused on gut microbiome associations with PRRS vaccine efficacy is limited to a single study showing no effect of an oral single-strain probiotic on PRRS vaccine response. However, the present disclosure provides specific methods of modulating the microbiome to improve weight gain and the efficacy of porcine vaccines.

SUMMARY

One aspect is a method of increasing weight gain in a pig, comprising modifying at least one aspect of the pig’s gut microbiome. Such method may comprise increasing the species diversity of the gut microbiome, which may comprise increasing the number of species in the gut microbiome to at least 50 to at least 80, or at least 82, at least 84, at least 86, at least 88, at least 90, at least 94, at least 98, at least 100, at least 105, at least 110, at least 115, at least 120, or at least 125 different species. Modifying at least one aspect of the pig’s gut microbiome comprises increasing the number of at least one bacterial species; the number of bacteria from at least one family; or, the number of bacteria from at least one phylum. The at least one bacterial species may be selected from the group consisting of Bacteroides pectinophilus , Lachnospiraceae species C6A11, Lachnospiraceae species P6B14, and a bacterial species of the phylum Spirochaetes. Increasing the number of at least one bacterial species; the number of bacteria from at least one family; or, the number of bacteria from at least one phylum may comprise administering to the pig a composition comprising Bacteroides pectinophilus , Lachnospiraceae species C6A11, Lachnospiraceae species P6B14, a bacterial species of the phylum Spirochaetes , or any combination thereof. Modifying at least one aspect of the pig’s gut microbiome may comprise modifying the Firmicutes/Bacteriodetes ratio in the gut microbiome to above 2:1, which may comprise modifying the Firmicutes/Bacteriodetes ratio in the gut microbiome to at least 2.2:1, at least 2.3:1, at least 2.4:1, at least 2.5:1, at least 2.6:1, at least 2.7:1, at least 2.8:1, at least 2.9:1, at least 3:1, at least 3.1:1, at least 3.2:1, at least 3.3:1, at least 3.4:1, at least 3.5:1, at least 3.6:1, at least 3.7:1, at least 3.8:1, at least 3.9:1, at least 4:1, at least 4:2:1, at least 4.4:1, at least 4.6:1, at least 4.8:1, at least 5:1, at least 5.5:1, at least 6:1, at least 6.5:1, at least 7:1, at least 8:1, or at least 9:1. In these methods, modifying at least one aspect of the pig’s gut microbiome comprises administering to the pig a composition comprising at least one bacteria. In some aspects, modifying at least one aspect of the pig’s gut microbiome may comprise reducing the Mycoplasmataceae species diversity in the gut microbiome, reducing the relative number of Mycoplasma bacteria, which may be Mycoplasma conjunctivae , in the gut microbiome or reducing the number of bacterial species in the family Lachnospiraceae , in the gut microbiome.

One aspect is a non-natural composition comprising at least one bacterial species selected from the group consisting of Bacteroides pectinophilus , Lachnospiraceae species C6A11, Lachnospiraceae species P6B14, and a bacterial species of the phylum Spirochaetes. The composition may comprise a stabilizer, a preservative, a buffer, a sugar, a protein, a salt, a mineral, a substrate for bacterial species, a coloring agent, a flavoring agent, or any combination thereof.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1 A & IB show weight gain in high and low growth rate pigs after PRRS MLV vaccination and co-challenge with PRRSV and PCV2b. FIG. 1 A shows average daily weight gains post challenge. The data are shown as mean ADG in kg ± one standard deviation for high and low growth rate groups post-challenge. FIG. IB shows weekly body weights during vaccination and challenge periods. Data are shown as the mean weight in kg ± one standard deviation with regression lines. Asterisks identify statistically significant differences in mean weights between the two groups (*p < 0.01, **p < 0.001 and ***p < 0.0001; two-way ANOVA with repeated measures).

FIGS. 2 A, 2B, 2C, and 2D show PRRSV and PCV2 virus replication in high and low growth rate pigs after PRRS MLV vaccination and PRRSV/PCV2 co-challenge. Data are shown as the mean logio copies/PCR reaction ± standard deviation for each virus (FIG2A shows PRRV viremia;

FIG. 2B shows PCV2b viremia) and growth rate group (FIGS. 2A and IB). On 32 dpv, the high growth rate group had a trend towards significantly higher PRRSV replication, while on 39 and 42 dpv, the low growth rate group had a trend towards significantly higher PRRS viremia. On 39 dpv, the high growth rate group had a trend towards significantly higher PCV2 viremia. Trends towards significance (ip < 0.1; repeated measures analysis using multiple t-tests) are highlighted. FIGS. 2C & 2D show the post-challenge total viral loads for each pig with horizontal lines representing mean ± one standard deviation. No significant differences in PRRSV or PCV2 viral load was noted post-challenge (unpaired /-test).

FIGS. 3A, 3B, 3C, 3D, 3E, and 3F show antibody response and lymphoid depletion in high and low growth rate pigs after PRRS MLV vaccination and co-challenge with PRRSV and PCV2. Data are shown as the mean Sample:Positive (S/P) ratio ± one standard deviation in each group for PRRSV (FIG. 3 A), PCV2 CP 43-233 epitope (FIG.3B), and PCV2 CP 160-233 decoy epitope FIG.3(C) on 28 and 63 dpv with p-values (unpaired t-tests). (FIG.3D) Data are shown as individual lymphoid depletion scores for each of the 20 pigs with horizontal lines representing mean scores ± one standard deviation. Representative histopathologic images of 2X H&E stained tonsils are shows as follows: score 0, no lymphoid depletion (FIG.3E) and score 3, severe lymphoid depletion (FIG.3F).

FIGS. 4A-4C show fecal microbiome diversity in pigs with high and low growth rates after PRRSV MLV vaccination and subsequent co-infection with PRRSV and PCV2. Data are shown as the total number of microbial families (FIG. 4A) and microbial species (FIG. 4B) detected by the LLMDA prior to co-challenge for individual pigs. Group means and standard deviations are represented by horizontal lines. No significant difference in microbiome diversity was detected on a family or species level between the two groups (p > 0.05; Mann-Whitney U test) (FIG. 4C) Data are shown as the Firmicutes:Bacteroidetes ratio for each pig in the two groups. Horizontal lines represent mean ± one standard deviation.

FIGS. 5 A & 5B show the fecal microbiome composition in pigs with high and low growth rates after PRRSV MLV vaccination and subsequent co-infection with PRRSV and PCV2. FIG. 5A shows the microbiome family prevalence as the percent of high growth (n = 10) and low growth (n = 10) rate pigs with each family detected on the LLMDA. Families detected in less than 40% of all 20 pigs are not shown. FIG. 5B shows the mean number of species detected ± one standard deviation in each family identified in high and low growth rate pigs. Statistical significance (*p < 0.05) and trends towards significance (ip < 0.1) are highlighted.

FIGS. 6A, 6B, and 6C show fecal microbiome analysis by 16S rDNA sequencing in high and low growth rate pigs. FIG. 6A shows the mean relative abundance of bacterial phyla for high and low growth groups. FIG. 6B shows the mean relative abundance of bacterial families for high and low growth groups. Families making up 1% or less of all sequences detected in 1 or more sample subsets are grouped together and classified as “Other”. FIG. 6C shows the a- diversity metric (Chaol or Shannon Index) for individual pigs in each group. Group means and standard deviations are represented by horizontal lines. Chaol a-diversity was significantly greater in the high growth group (Mann Whitney U test).

DETAILED DESCRIPTION

The present disclosure relates to methods of increasing weight gain in a pig, either prior to or following infection of the animal. Such methods involve modifying one or more aspects of the gut microbiome of the pig. Thus, a method of the disclosure may generally be practiced by modifying at least one aspect of the gut microbiome of the pig. Such modification may include increasing or decreasing the relative number of specific bacteria, increasing the diversity of bacteria, or increasing the ratio of particular bacterial species, in the gut microbiome. Accordingly, the present disclosure also relates to compositions for modifying the number, diversity, and/or ratio of bacteria in the gut microbiome of a pig.

Before the present disclosure is further described, it is to be understood that this disclosure is not limited to particular embodiments described, as such may, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting, since the scope of the present disclosure will be limited only by the claims.

It must be noted that as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. For example, a compound refers to one or more compound molecules. As such, the terms “a”, “an”, “one or more” and “at least one” can be used interchangeably. Similarly, the terms “comprising”, “including” and “having” can be used interchangeably. It is further noted that the claims may be drafted to exclude any optional element. As such, this statement is intended to serve as antecedent basis for use of such exclusive terminology as “solely,” “only” and the like in connection with the recitation of claim elements or use of a “negative” limitation.

One aspect of the disclosure is a method of increasing weight gain in a pig, comprising modifying at least one aspect of the pig’s gut microbiome. As used herein, “pig”, “hog”, “boar”, and the like, refer to an animal in the family Suidae. The term “pig” encompasses both male and female animals and encompasses any age of animal. In some aspects, the pig may be from the genus Sus, which includes wild and domestic pig. In some aspects, the pig may be a domestic pig ( Sus domesticus). The term “pig” also encompasses any breed of Sus, examples of which include, but are not limited to, American Yorkshire, Berkshire, Chester White, Duroc,

Hampshire, Hereford, Landrace, Large Black, Spotted, Gloucestershire Old Spot, Pietrain,

Poland China, and Tamworth.

As used herein, “modifying at least one aspect of’ a microbiome means altering the number of bacterial families or species in the microbiome. Such altering may include, for example, increasing or decreasing the number of one or more particular species in the microbiome, the ratio of two or more bacterial families or species in the microbiome, or increasing the diversity of bacteria in the microbiome. The diversity of bacteria refers to the number of different families or species from a family, of bacteria in the microbiome.

The term “gut” refers to the organs, glands, tracts, and systems that are responsible for the transfer and digestion of food, absorption of nutrients, and excretion of waste. The gut comprises the gastrointestinal (GI) tract, which starts at the mouth and ends at the anus, and additionally comprises the esophagus, stomach, small intestine, and large intestine. The gut also comprises accessory organs and glands, such as the spleen, liver, gallbladder, and pancreas. The upper gastrointestinal tract comprises the esophagus, stomach, and duodenum of the small intestine. The lower gastrointestinal tract comprises the remainder of the small intestine, i.e., the jejunum and ileum, and alle of the large intestine, i.e., the cecum, colon, rectum, and anal canal. Bacteria are be found throughout the gut, e.g., in the gastrointestinal tract, and particularly in the intestines.

As used herein, the term “microbiome” refers to the community of microorganisms that inhabit (sustainably or transiently) in and/or on a subject, (e.g, a mammal such as a pig), including, but not limited to, eukaryotes (e.g., protozoa), archaea, bacteria, and viruses (including bacterial viruses, i.e., a phage). As such, “gut microbiome” refers to the community of microorganisms that inhabit, sustainably or transiently, the gut of a subject. The terms “subject” and “individual” may be used interchangeably with regard to “pig”.

In some aspects of the disclosure, a method for increasing weight gain in pigs may comprise increasing the species diversity in the microbiome. In some aspects, the number of species in the microbiome is increased to at least 50 different species, at least 51 different species, at least 52 different species, at least 53 different species, at least 54 different species, at least 55 different species, at least 56 different species, at least 57 different species, at least 58 different species, at least 59 different species, at least 60 different species, at least 61 different species, at least 62 different species, at least 63 different species, at least 64 different species, at least 65 different species, at least 66 different species, at least 67 different species, at least 68 different species, at least 69 different species, at least 70 different species, at least 71 different species, at least 72 different species, at least at least 73 different species, at least 74 different species, at least 75 different species, at least 76 different species, at least 77 different species, at least 78 different species, at least 79 different species, at least 80 different species, at least 82 different species, at least 84 different species, at least 86 different species, at least 88 different species, at least 90 different species, at least 94 different species, at least 98 different species, at least 100 different species, at least 105 different species, at least 110 different species, at least 115 different species, at least 120 different species, or at least 125 different species. In some aspects, the number of species in the microbiome is increased so that it is greater than 125 species.

In some aspects of the disclosure, a method for increasing weight gain in pigs may comprise increasing the number of at least one bacterial species, or the number of bacteria from at least one family or at least one phylum. Such a method may be accomplished by administering to a pig a composition comprising at least one bacterial species and/or bacteria from at least one family or at least one phylum.

In some aspects of the disclosure, a method for increasing weight gain in pigs may comprise increasing the number of at least one bacterial species selected from the group consisting of Bacteroides pectinophilus , Lachnospiraceae species C6A11 and Lachnospiraceae species P6B14. In some aspects, such a method may be accomplished by administering to a pig a composition comprising at least on bacterial species selected from the group consisting of Bacteroides pectinophilus, Lachnospiraceae species C6A11 and Lachnospiraceae species P6B14.

In some aspects of the disclosure, a method for increasing weight gain in pigs may comprise modifying the Firmicutes/Bacteroidetes ratio so that it is above 2:1. Firmicutes refers to bacteria of the phylum Firmicutes , which may also be referred to as Bacillota. Bacteroidetes refers to bacteria of the phylum Bacteroidetes. The Firmicutes/Bacteroidetes ratio refers to the ratio of the number, or percentage, of bacteria, or species, from the phylum Firmicutes to the number, or percentage, of bacteria, or species, from the phylum Bacteroidetes. Methods of determining the Firmicutes/Bacteroidetes ratio are disclosed herein and are also known in the art. In some aspects, the gut microbiome is modified so that the Firmicutes/Bacteroidetes ratio is at least 2.1:1, at least 2.2:1, at least 2.3:1, at least 2.4:1, at least 2.5:1, at least 2.6:1, at least 2.7:1, at least 2.8:1, at least 2.9:1, at least 3:1, at least 3.1:1, at least 3.2:1, at least 3.3:1, at least 3.4:1, at least 3.5:1, at least 3.6:1, at least 3.7:1, at least 3.8:1, at least 3.9:1, at least 4:1, at least 4:2:1, at least 4.4:1, at least 4.6:1, at least 4.8:1, at least 5:1, at least 5.5:1, at least 6:1, at least 6.5:1, at least 7:1, at least 8:1, at least 9:1, or at least 10:1. In some aspects, the gut microbiome is modified so that the Firmicutes/Bacteroidetes ratio is greater than 10:1.

In some aspects of the disclosure, a method for increasing weight gain in a pig may comprise the steps of increasing the number of bacteria from the phylum Spirochaetes. In some aspects, such a method may comprise increasing the number of species of bacteria from the phylum Spirochaetes. In some aspects, such a method may comprise increasing the number all species of the phylum Spirochaetes, or increasing one or more specific species of the phylum Spirochaetes. In some aspects, such a method may be accomplished by administering to a pig a composition comprising one or more species of bacteria from the phylum Spirochaetes .

In some aspects of the disclosure, a method for increasing weight gain in pigs may comprise decreasing the diversity of Mycoplasmataceae species in the gut microbiome.

In some aspects of the disclosure, a method for increasing the weight gain in a pig may comprise decreasing the overall number of Mycoplasma conjunctivae in the gut microbiome.

In some aspects of the disclosure, a method for increasing the weight gain in a pig may comprise decreasing the overall number of bacteria from the family Lachnospiraceae .

In some aspects, weight gain may be determined by average daily weight gain.

In some aspects, modifying the gut microbiome may comprise enhancing (adding bacteria to) or reducing (removing bacteria from) the gut microbiome naturally present in the pig. In some aspects, modifying the gut microbiome may comprise reducing or eliminating some or all of the bacteria present in the natural microbiome present in a pig, and administering to the pig a composition comprising a desired number, type, and/or ratio of bacteria. Reduction or elimination of bacteria in the natural microbiome may be accomplished, for example, using anti bacterial compounds (e.g., antibiotics).

Administration of a composition comprising bacteria for the purpose of increasing weight gain in pigs may be accomplished by any means suitable for modifying the gut microbiome.

Such means include, but are not limited to, oral administration, adding a composition to feedstuffs, adding a composition to a source of water, administration using an enteral feeding tube, and administration as a pill or capsule.

In some aspects, modification of the gut microbiome of a pig is performed prior to a point in time when the pig is infected by an organism. In some aspects, modification of the gut microbiome of a pig is performed following a point in time when the pig is infected. In some aspects, aspects, the organism may be porcine reproductive and respiratory syndrome (PRRSV) or porcine circovirus type 2 (PCV2).

In some aspects, methods of the disclosure may be used in pigs that have been vaccinated for PRRSV and/or PCV2. In some aspects, methods of the disclosure may be used to enhance the efficacy of a vaccine for PRRSV and/or PCV2.

One aspect of the disclosure is a non-natural composition comprising at least one bacterial species selected from the group consisting of Bacteroides pectinophilus , Lachnospiraceae species C6A11 , Lachnospiraceae species P6B14, and Ruminococcaceae bacterium AE2021. As used herein, a non-natural composition is a composition that has been created by the hand of man. While such a composition may comprise microorganisms and compounds obtained from nature, the microorganisms and/or are first isolated from nature by the hand of man before being used to produce the non-natural composition. As used herein, the term “isolated” means a microorganism or compound that has been separated from its natural environment. As such, “isolated” does not denote a particular degree of purity. Bacteriia used to produce a non-natural composition of the disclosure may also be obtained from cultures of bacteria. In some aspects, a non-natural composition of the disclosure may comprise other ingredients to stabilize the composition, or aid in delivery or survival of the components in the composition. Examples of such additional ingredients include, but are not limited to, stabilizers, preservatives, buffers, sugars, proteins, salts, minerals, substrates for bacterial species, coloring agents, and flavoring agents. One aspect of the disclosure is a non-natural composition comprising at least one bacterial species of the phylum Spirochaetes. The non-natural composition may comprise at least one additional ingredient, such as a stabilizer, a preservative, a buffer, a sugar, a protein, a salt, a mineral, a substrate for bacterial species, a coloring agent, a flavoring agent, or any combination thereof.

EXAMPLES

Methods

Animals and housing

All use and experimentation of animals and viruses were done in accordance with the Federation of Animal Science Societies (FASS) Guide for the Care and Use of Agricultural Animals in Research and Teaching, the USD A Animal Welfare Act and Animal Welfare Regulations, and approved by the Kansas State University Institutional Animal Care and Use Committee and Institutional Biosafety Committee. This study was conducted as a part of the PRRS Host Genetics Consortium (PHGC) as described previously (Lunney et al., BMC Proceedings 5, S30, 2011). A subset of barrows ( n = 50; mean age 23.4 ± 2.1 days) were obtained at weaning from a high health commercial herd negative for PRRSV. Piglets were not vaccinated for PCV2 and were used without regards to maternal antibody. Piglets were housed in a single environmentally controlled room at the Kansas State University Large Animal Research Center and maintained under BSL-2 conditions. Piglets were randomly distributed into six 13.4 m ² pens, and housed in groups of 7-10 pigs per pen. Pigs were given access to food and water ad libitum.

Viruses

The PRRSV and PCV2b viral isolates used in this study originated from the lymph node of a pig with severe postweaning multisystemic wasting syndrome (PMWS) as previously described (Trible et al., 2011, CVI 18, 749-757; Trible et al., Vaccine 30, 4079-4085, 2012). PRRSV (isolate KS62; GenBank accession no. KM035803) was isolated by propagation on MARC-145 cells and PCV2b (GenBank accession no. JQ692110) was isolated by utilizing the heat stability of the virus and preparing a lymph node suspension enriched for PCV2. The procedures used to isolate and titrate the viruses have been described previously in detail (Niederwerder et al., Clinical and vaccine immunology : CVI 22, 1244-1254, 2015; Niederwerder et al., Veterinary microbiology 188, 1-11, 2016; Ober et al., Veterinary microbiology 208, 203-211, 2017). Analysis of the PCV2b tissue homogenate used for challenge detected two ubiquitous swine viruses, including torque teno sus virus (TTSuV) and porcine endogenous retrovirus (PERVs) (Jaing et al., 2015). PRRSV was quantified on MARC-145 cells and swine testicle cells were used to quantify PCV2. The 50% tissue culture infectious dose per milliliter (TCTD o/mL) was calculated using the Reed and Muench method (Reed and Muench, The American Journal of Hygiene 27, 493-497, 1938).

Experimental design and sample collection

After 4 days of acclimation, all 50 pigs were vaccinated with a 2-ml dose of a commercial PRRS MLV vaccine (Ingelvac PRRS MLV; Boehringer Ingelheim Animal Health, Duluth, GA; GenBank accession no. AF159149). The vaccine was administered intramuscularly according to the vaccine label instructions. At 28 days post-vaccination (dpv) and approximately 8 weeks of age (mean age 55.4 ± 2.1 days), all pigs were challenged with a combination of PRRSV and PCV2b. The challenge viruses were combined to yield a 2-ml dose consisting of 10 ³⁶ TCIDso PCV2b and 10 ⁵ TCIDso PRRSV in MEM. The 2-ml dose was split, with 1 ml administered intranasally and 1 ml administered intramuscularly. Rationale for co-infection has been described in detail (Niederwerder, Veterinary microbiology 209, 97-106, 2017) and includes the prevalence of PRRSV as a contributor to PCVAD and the potentiation of immunomodulation.

Individual body weights were determined on 0, 7, 14, 21, 28, 35, 42, 49, 56, 63, and 70 dpv. Blood samples were collected on 0, 4, 7, 11, 14, 21, 28, 32, 35, 39, 42, 49, 56, 63, and 70 dpv. Fecal samples were collected from all 50 pigs during the week prior to co-challenge for microbiome analysis. At 42 days post-challenge (dpc) or 70 dpv, 20 pigs were retrospectively identified as having high (n = 10) or low (// = 10) growth rates during the co-challenge period.

To select these groups, average daily gain (ADG) was calculated as the change in weight over the 42-day period and reported in kg. Any pig displaying clinical signs which required veterinary medical treatment (as described below) were excluded from the study. At 42 dpc, all 20 pigs were humanely euthanized in accordance with the American Veterinary Medical Association Guidelines for the Euthanasia of Animals and complete necropsies were performed.

Clinical and Histological Evaluation

Pigs were visually examined by a veterinarian or veterinary assistant on each day of the study for clinical signs associated with PRRSV/PCV2 co-challenge, including respiratory signs (e.g., dyspnea, coughing, nasal discharge, ocular discharge and open mouth breathing), lethargy, depression, diarrhea, pyrexia, lameness, joint effusion, decreased body condition, muscle wasting and aural cyanosis. Any pig showing moderate to severe clinical disease was treated or euthanized under the direction of a veterinarian. Examples of clinical signs in which treatment was administered included 1) dyspnea, 2) mucopurulent nasal discharge, 3) lameness with associated joint effusion, 4) pallor with muscle wasting, and 5) lethargy or depression with pyrexia. Clinically affected pigs were administered parenteral antibiotics, including ceftiofur hydrochloride (Excede®; Zoetis, Parsippany, NJ), oxytetracycline (Liquamycin® LA-200®; Zoetis, Parsippany, NJ), or enrofloxacin (Baytril®; Bayer Healthcare LLC, Shawnee Mission, KS). Pigs with overt clinical disease and rectal temperatures of > 104°F were administered a nonsteroidal anti-inflammatory drug, such as flunixin meglumine (Banamine®; Merck Animal Health, Madison, NJ) or meloxicam. Any pig with documented clinical disease requiring veterinary medical treatment was excluded from selection for fecal microbial analysis.

At 42 dpc, all pigs were humanely euthanized using pentobarbital sodium (Fatal -Plus®; Vortech Pharmaceuticals, Dearborn, MI). A masked board-certified veterinary pathologist performed complete necropsies and histopathologic evaluations. Tonsils were collected and fixed in 10% neutral buffered formalin for at least 7 days, routinely processed in an automated tissue processor, embedded in paraffin, and stained with hematoxylin and eosin (H&E stain).

Lymphoid depletion was scored on a scale of 0-3 as previously described (Niederwerder et ak, 2016). Briefly, scores were given as follows: 0, no lymphoid depletion; 1, mild or small amount of lymphoid depletion; 2, moderate or intermediate amount of lymphoid depletion; 3, severe or large extent of lymphoid depletion.

Measurement of PRRSV and PCV2 viremia and viral load

Methods utilized to measure PRRSV and PCV2 viremia have been described in detail previously (Niederwerder et ak, Clinical and vaccine immunology : CVI 22, 1244-1254, 2015; Niederwerder et al., Veterinary microbiology 188, 1-11, 2016; Ober et al., Veterinary microbiology 208, 203-211, 2017). Briefly, viral DNA and RNA were extracted simultaneously from 50 pL of serum using Ambion’s MagMAX 96 Viral Isolation Kit (Applied Biosystems, Thermo Fisher Scientific, Waltham, MA) in accordance with the manufacturer’s instructions. PRRS viral RNA was quantified using EZ-PRRSV MPX 4.0 Real Time RT-PCR Target-Specific Reagents (Tetracore, Rockville, MD) according to the manufacturer’s instructions. The PRRSV PCR assay results were reported as logio RNA starting quantity (copy number) per 50-pl reaction volume. PCV2 DNA was quantified using SsoAdvanced Universal SYBR green supermix (Bio- Rad, Hercules, CA). The forward and reverse PCR primers were 5’-

AATGCAGAGGCGTGATTGGA-3 ’ (SEQ ID NO.5) and 5’-CCAGTATGTGGTTTCCGGGT- 3’ (SEQ ID NO.6), respectively. Standard curves and positive and negative controls were included on each plate. The PCV2b PCR assay results were reported as logio DNA starting quantity (copy number) per 20 pL reaction volume. Total viral load for PRRSV and PCV2 were calculated by Riemann sums of the total area of the trapezoids under the line segments connecting weekly or biweekly viremia measurements. Viral load was calculated during the vaccination period (0-28 dpv) for PRRSV MLV vaccine replication and during the challenge period (28-70 dpv) for PRRSV and PCV2 replication.

Microsphere immunoassay for detection of PRRSV and PC 2 antibodies

PRRSV nucleocapsid protein and PCV2b capsid protein polypeptide fragments (43-233 and 160-233) were cloned into the pHUE vector, as previously described in detail (Niederwerder et al., 2018). Proteins were expressed, purified and measured prior to being coupled to carboxylated Luminex MagPlex ^® polystyrene microspheres (Luminex Corporation, Austin, TX) according to the manufacturer’s instructions. For the assays, approximately 2500 antigen-coated beads, suspended in 50 pL PBS with 0.05% Tween-20 and 4% goat serum (PBST-GS), were placed in each well of a 96-well polystyrene round bottom plate (Coming® Costar®

Corporation, Cambridge, MA). Sera were diluted 1 :400 in PBST-GS and 50 pL was added to each well. An adhesive foil plate sealer was applied and the plate was incubated for 30 min at room temperature with gentle shaking. After incubation, the plate was placed on a magnet and beads were washed three times with 190 pL of PBS-GS. For the detection of IgG, 50 pL of biotin-SP-conjugated affinity purified goat anti-swine secondary antibody (IgG, Jackson ImmunoResearch, West Grove, PA) was diluted to 2 pg/mL in PBST-GS and 100 pL was added to each well. The plate was incubated at room temperature for 30 min and washed three times followed by the addition of 50 pL of streptavi din-conjugated phycoerytbrin (2 ug/mf in PBST- GS; SAPE). After 30 min, the plate was washed and microspheres resuspended in 100 pL of PBST-GS. Microspheres were analyzed using a MAGPIX instrument (Luminex Corporation, Austin, TX) and Luminex ^® xPONENT 4.2 software. A minimum of 50 microspheres was used for the calculation of mean fluorescence intensity (MFI). The sample to positive (S/P) ratio was calculated as the MFI of the sample minus the MFI of the negative control divided by the MFI of the standard positive control minus the MFI of the negative control.

Microarray Analysis of fecal microbiome

The Lawrence Livermore Microbial Detection Array (LLMDA) was used to analyze microbiome composition and diversity of fecal samples collected prior to co-challenge. This specific array detects annotated sequences of microbes within GenBank®, the National Institute of Health genetic sequence database. The version 7 of the LLMDA in the 4plex 180K probe format was used in this study, which detects a total of 10,612 microorganisms including 5,457 bacteria, 4,377 viruses, 327 archaebacteria, 319 fungi, and 132 protozoa. Probe lengths on the array is around 60 nt and have roughly equivalent affinities for their complementary target DNA molecules (McLoughlin, Briefings in functional genomics 10, 342-353, 2011). Probes were designed to detect all sequenced microbial families with a large number of probes per sequence (average of 30 probes) to improve sensitivity. The high-density oligo LLMDA microarray and statistical analysis method have been extensively tested in numerous studies for viral and bacterial detection in pure or complex environmental and clinical samples (Gardner et al., BMC genomics 11, 668, 2010; Jaing et al., Journal of veterinary diagnostic investigation : official publication of the American Association of Veterinary Laboratory Diagnosticians, Inc 27, 313- 325, 2015; Niederwerder et al., Clinical and vaccine immunology : CVI 22, 1244-1254, 2016; Rosenstieme et al., PloS one 9, el00813, 2014).

The PowerViral™ Environmental RNA/DNA Isolation Kit (MO BIO, San Diego, CA) was used to extract DNA and RNA from the fecal samples. For each sample, approximately 0.25 g of feces was added to 600 mΐ of Rnΐ/b-mercaptoethanol in a glass bead tube included in the kit. Samples were homogenized and lysed by vortexing tubes for 10 minutes at maximum speed. Samples were further processed using the PowerViral™ Kit protocol. All samples were eluted into 100 mΐ of RNase-Free water. The purified nucleic acids were quantified using the Thermo Scientific™ Nanodrop™ spectrophotometer (Thermo Fisher Scientific, Waltham, MA). For each sample, 10 mΐ of the extracted DNA and RNA was amplified using the random amplification procedure as previously described (Rosenstierne et al., 2014). The amplified cDNA and DNA was purified with the Qiaquick PCR purification columns (Qiagen, Hilden, Germany) and quantified using the Nanodrop™ spectrophotometer.

Approximately 1 pg of amplified cDNA and DNA were fluorescently labeled using a one-coloring labeling protocol (Roche NimbleGen, Madison, WI). Briefly, the samples were labeled using nick translation with Cy3 -labeled random nonamer primers (TriLink Biotechnologies, San Diego, CA) and Klenow DNA polymerase (New England Biolabs,

Ipswich, MA) at 37°C for 2 hr. The labeled DNA was precipitated in isopropanol, and the pellet was washed, and dried. The pellet was then reconstituted in 50 mΐ of RNase-Free water and quantified using the Nanodrop™ spectrophotometer.

The Agilent Technologies Oligo aCGH/ChIP-on-Chip Hybridization kit (Santa Clara,

CA) was used to hybridize samples to the arrays. For each sample, 10 pg of fluorescently labeled DNA was mixed with lOx aCGH blocking agent, 2x HiRPM hybridization buffer and nuclease free water. The samples were then denatured at 95 °C for 3 min, and incubated at 65 °C for 3 min. Each sample was then immediately loaded onto the array and hybridized for approximately 40 hr at 65°C in a microarray rotator oven (Agilent Technologies Inc., Santa Clara, CA) set to a speed of 20. Microarrays were washed using the standard manufacturer’s protocol with Oligo aCGH/ChIP-on-chip Wash Buffer 1 for 5 min at room temperature and Oligo aCGH/ChlP-on- chip Wash Buffer 2 for 1 min at 37°C (Agilent Technologies Inc., Santa Clara, CA). Using the SureScan Microarray Scanner (Agilent Technologies Inc., Santa Clara, CA), all arrays were scanned to a resolution of 3 pm.

Microarray data was generated from the microbe sequences using the CLiMax method developed at Lawrence Livermore National Laboratory, at a detection threshold of > 99%. The log likelihood for each of the positive targets is estimated from the BLAST similarity scores of the array feature and target sequences, together with the feature sequence complexity and other covariates derived from BLAST results. Diversity of the fecal samples was measured by calculating the number of families and species detected in each sample. The mean number of families and species were compared between high and low growth rate groups. Microbiome composition was compared between the two groups at the level of phylum, family and species. The Firmicutes to Bacteroidetes (Firmicutes/Bacteroidetes) ratio was determined by dividing the total number of Fimicutes bacterial species by the total number of Bacteroidetes bacterial species detected for each pig.

16S rDNA Analysis of Fecal Microbiome

DNA was extracted from fecal samples as described above for microarray detection analysis prior to submission to the University of Minnesota Genomics Center (UMGC) for 16S library preparation of the V4 region using standard diagnostic protocols and a two-step PCR protocol as described previously (Gohl et al., Nature biotechnology 34, 942-949, 2016). Briefly, a dual -indexing protocol was utilized that uses a single pair of PCR primers with 5' adaptor tails to amplify samples in a 'primary ¹ amplification, while a 'secondary' PCR adds flow cell adaptors and indices.

The primary amplification was done in a qPCR reaction, using AB 17900. The following recipe was used: 3 pi of template DNA, 0.48 mΐ of nuclease-free water, 1.2 mΐ of 5x KAPA HiFi buffer (KAPA Biosystems, Woburn, MA), 0.18 mΐ of 10 mM dNTPs (KAPA Biosystems, Woburn, MA), 0.3 mΐ of DMSO (Fisher Scientific, Waltham, MA), 0.12 mΐ of ROX (25 mM)

(Life Technologies, Carlsbad, CA), 0.003 mΐ of l,000x SYBR Green, 0.12 mΐ of KAPA HiFi polymerase (KAPA Biosystems, Woburn, MA), 0.3 mΐ of forward primer (10 mM), and 0.3 mΐ of reverse primer (10 mM). Cycling conditions were as follows: 95°C for 5 min, followed by 35 cycles of 98°C for 20 s, 55°C for 15 s, and 72°C for 1 min. The primers for the primary amplification contained both 16S-specific primers (V4 515F and V4 806R) and adaptor tails for adding indices and Illumina flow cell adaptors in a secondary amplification. The following primers were used (16S-specific sequences in bold):

V4_515F_Nextera:

TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGGTGCCAGCMGCCGCGGTAA

(SEQ ID NO:!) V4_806R_N exter a :

GTCTC GT GGGC T C GGAGAT GT GT AT A AGAGAC AGGGACT ACH V GGGT WTCT A AT (SEQ ID NO:2)

The amplicons from the primary PCR were diluted 1 : 100 in sterile, nuclease-free water, and a second PCR reaction was set up to add the Illumina flow cell adaptors and indices. The secondary amplification was done using the following recipe: 5 pi of template DNA, 1 mΐ of nuclease-free water, 2 mΐ of 5x KAPA HiFi buffer (KAPA Biosystems, Woburn, MA), 0.3 mΐ of 10 mM dNTPs (KAPA Biosystems, Woburn, MA), 0.5 mΐ of DMSO (Fisher Scientific, Waltham, MA), 0.2 mΐ of KAPA HiFi Polymerase (KAPA Biosystems, Woburn, MA), 0.5 mΐ of forward primer (10 mM), and 0.5 mΐ of reverse primer (10 mM). Cycling conditions were as follows: 95°C for 5 min; ten cycles of 98°C for 20 s, 55°C for 15 s, and 72°C for 1 min; and a final extension at 72°C for 10 min. The following indexing primers were used (p5 and p7 flow cell adapters are in bold; X indicates the positions of the 8-bp Illumina indices):

1) Forward indexing primer:

AATGATACGGCGACCACCGAGATCTACACNNNNNNNNTCGTCGGCAGCGTC (SEQ ID NO:3)

2) Reverse indexing primer:

CAAGCAGAAGACGGCATACGAGATNNNNNNNNGTCTCGTGGGCTCGG (SEQ ID NO:4)

Products were quantified using a PicoGreen dsDNA assay kit (Life Technologies, Carlsbad, CA), normalized and pooled the samples, and concentrated approximately 1 pg of material to 10 mΐ using 1.8* AMPureXP beads (Beckman Coulter, Brea, CA). The pooled sample was then size-selected at 427 bp ± 20%, on a Caliper XT DNA 750 chip (Caliper Life Science, Hopkinton, MA). The size-selected material was cleaned up using AMPureXP beads and eluted in 20 mΐ of EB buffer (10 mM Tris-HCl, pH 8.5). The final pooled sample was quantified using the PicoGreen dsDNA assay. The sample pools were diluted to 2 nM on the basis of the PicoGreen measurements, and 10 mΐ of the 2 nM pool was denatured with 10 mΐ of 0.2 N NaOH, diluted to 8 pM in Illumina's HT1 buffer, spiked with 20% PhiX, heat-denatured at 96 °C for 2 min and immediately sequenced with a MiSeq 600 cycle v3 kit (Illumina, San Diego, CA). For data processing, UMGC’s bioinformatics pipeline which implements QIIME (Caporaso et al., Nature Methods 7, 335, 2010) version 1.9.1 analysis software was used. Raw fastq files were filtered for primer and adapter dimer sequences, removing contaminating host sequences and chimeric sequences, clustering sequences into OTUs using the QIIME open- reference OTU calling method with the greengenes 16s reference. Sequencing adapter sequences were trimmed from the 3’ ends of reads using Trimmomatic (Bolger et al., Bioinformatics (Oxford, England) 30, 2114-2120, 2014). PandaSeq (version 2.7) (Masella et al., BMC Bioinformatics 13, 31, 2012) was used to remove primer sequences from the beginning of reads and to stitch the overlapping paired reads together. Reads without primer sequences and reads that could not be stitched together were discarded. Stitched reads whose length were outside the expected length of the targeted variable region were discarded. FASTQ files were converted to QIIME FASTQ format using a custom script. Individual sample FASTA files were concatenated into one FASTA file, chimera detection was run and chimeric sequences were removed using ChimeraSlayer's usearch61 method (Haas et al., Genome Research 21, 494-504, 2011). Contaminating host sequences were identified and discarded by aligning stitched reads to the HOST reference genome using BWA/BOWTIE2 [if a bwa or bowtie2 index was provided to the pipeline]. For OTU picking, we used QIIME's pick open reference otus.py script with usearch61. Taxonomy was assigned using QIIME's assign taxonomy.py script, which uses the RDP classifier and the Greengenes reference database clustered at 97% identity. Reference-based OTUs were then collapsed according to taxonomy at the genus level using QIIME's summarize taxa.py script.

Statistical Analysis

Statistical analysis of 16S rDNA data was conducted with the Dynamic Assessment of Microbial Ecology (DAME) server (Piccolo et al., 2018). DAME is an open source platform that uses the R environment to analyze and visualize microbial sequencing data. Alpha (a-) and beta (b-) diversity as well as relative abundance were analyzed between high and low growth rate groups. The a-diversity and b-diversity indices were calculated based on the rarefied OTU counts. A Mann-Whitney U test was used to compare a-diversity between the two groups; specifically Chaol was used to determine species richness and the Shannon’s index was used to determine species evenness. Permutational multivariate analysis of variance (PERMANOVA) was performed on b-diversity measures using the adonis2() function from the vegan package with a setting of 500 permutations. Bray-Curtis analysis was used to analyze differences between the two groups. Negative binomial regression (McMurdie and Holmes, PLoS computational biology 10, el003531-el003531, 2014) using the R package DESeq2 (Love et ah, Genome biology 15, 550-550, 2014) was used to determine differential abundance of individual taxa. Likelihood ratio tests for overall experimental group comparisons was used to compare differential abundance between the two groups.

All remaining statistical analyses were performed using GraphPad Prism® 7.01 software (La Jolla, CA). Mean weekly weights was measured by repeated measures two-way ANOVA. ADG, mean viremia, viral load, lymphoid depletion, antibody levels, the mean Firmicutes/Bacteroidetes ratio, and the mean number of Proteobacteria were compared between groups using the unpaired /-test. Mean arrival age, mean microbiome diversity and mean number of species within each family were compared between groups using the Mann-Whitney U test. Proportions of each group with individual species and families detected were compared using Fisher’s exact test.

Results

Weight gain divergence after PRRS vaccination and PRRSV/PCV2 co-challenge led to groups of high and low growth rate pigs

Of the 50 pigs that had fecal samples collected prior to co-challenge, twelve pigs were excluded from the microbiome study due to the presence of clinical disease and/or mortality. Therefore, 38 pigs were considered for selection into high and low growth rate groups for fecal microbiome analysis. These pigs supported subclinical infections or had mild and transient clinical disease that did not require veterinary intervention or antimicrobial therapy. ADG between 0 and 42 dpc was utilized as the selection criteria for the two groups in this study (Figure 1 A). Mean ADG for the high growth rate group was 0.95 ± 0.06 kg, with a range of 0.87 kg and 1.03 kg. Mean ADG for the low growth rate group was 0.71 ± 0.11 kg, with a range of 0.49 kg and 0.83 kg. No overlap occurred between the ADG values of individual pigs within the two groups and the mean ADG was significantly different over the 42 day co-challenge period between the two groups (p < 0.001; unpaired t-test). Mean weekly weights between the two groups were similar during the vaccination period (p > 0.05; repeated measures two-way ANOVA); however, ADG over the entire 28-day vaccine period was significantly greater in the high growth rate group. Mean ADG prior to challenge was 0.47 ± 0.08 kg and 0.36 ± 0.09 kg for the high and low growth rate groups, respectively (p = 0.008; unpaired t-test). A significant divergence in the mean absolute weekly weights of the two groups occurred at 7 dpc; mean weights of 25.5 ± 3.8 kg and 19.5 ± 3.4 kg were measured in high and low growth rate groups, respectively (p = 0.01; repeated measures two-way ANOVA). Weekly weights continued to be significantly different between the two growth rate groups for the remaining 5 weeks of the post challenge period (Figure IB). Over the course of the 70 day study, the growth rate slopes of the two groups were significantly different (p = 0.004; linear regression). At the conclusion of the study, the high growth rate group weighed an average of 59.2 kg compared to 44.6 kg in the low growth rate group (p < 0.0001; repeated measures ANOVA). The slope of weight gain in high growth rate pigs was significantly greater than low growth rate pigs (p = 0.004; linear regression analysis).

Viremia, antibody production and lymphoid depletion of high and low growth rate groups

Virus replication on weekly to bi-weekly serum sampling days and total viral load in serum were determined for both PRRSV and PCV2b. PRRSV viremia had a bimodal distribution with peaks associated with vaccine virus replication and subsequent challenge virus replication (Figure 2A). Vaccine virus replication was similar between the two growth rate groups, with peak MLV replication occurring at 11 dpv in both groups (p > 0.05; repeated measures analysis). After challenge with wildtype PRRSV, high growth rate pigs had a more rapid incline in virus replication followed by a more rapid decline compared to low growth rate pigs. On 32 dpv, mean PRRSV viremia was 3.1 and 2.5 logio copies/PCR reaction for the high and low growth rate groups, respectively (p = 0.086; repeated measures analysis using multiple t-tests). High growth rate pigs peaked PRRS challenge virus replication at 35 dpv whereas low growth rate pigs peaked 4 days later at 39 dpv. At 39 dpv, there was a trend towards a significant reduction in the PRRS viremia of high growth rate pigs, with a mean of 2.6 logio copies/PCR reaction being detected compared to 3.2 logio copies/PCR reaction in the low growth rate group (p = 0.076; repeated measures analysis using multiple t-tests). This trend continued at 42 dpv, where high growth rate pigs (mean 2.1 logio copies/PCR reaction) had reduced PRRSV detected in serum compared to the low growth rate group (2.7 logio copies/PCR reaction; p = 0.091, repeated measures analysis using multiple t-tests). Overall, PRRSV viral load during the vaccination and challenge periods were similar between the two growth rate groups (Figure 2C; p = 0.78 and p = 0.95, respectively; unpaired /-test).

PCV2 viremia curves were similar between the two growth rate groups post-challenge, with peak virus replication occurring at 49 dpv followed by a generalized plateau of continued virus detection in the serum until 70 dpv (Figure 2B). On 39 dpv, there was a trend towards high growth rate pigs having higher PCV2 viremia compared to low growth rate pigs (p = 0.054; repeated measures analysis using multiple t-tests); PCV2 detection was approximately 1 logio copies/PCR reaction greater in high growth rate pigs. Overall, PCV2 total viral loads were similar between the two groups; 57.0 ± 35.5 for high growth pigs and 46.9 ± 29.9 for low growth pigs (Figure 2D; p = 0.50; unpaired /-test).

Antibody production against PRRSV N protein, PCV2 whole capsid protein (CP 43-233), and PCV2 decoy epitope (CP 160-233) were quantified at 28 dpv (prior to co-challenge) and 63 dpv (35 dpc; Figures 3A-C). Although antibody levels were greater numerically in the high growth rate group against PRRSV N protein after vaccination and after challenge, no significant differences were detected (p > 0.1; unpaired t-test). Antibody levels directed at the PCV2 capsid protein were also similar between the two groups. However, low growth rate pigs had significantly higher levels of baseline antibodies directed against PCV2 whole capsid protein (CP 43-233); 0.12 ± 0.05 S:P ratio in high growth pigs compared to 0.27 ± 0.20 S:P ratio in low growth pigs (p = 0.04; unpaired /-test). These low levels of detectable antibodies prior to challenge are likely associated with passive maternal transfer while nursing.

Tonsillar tissues were examined for lymphoid depletion associated with porcine circovirus associated disease at 70 dpv and compared between high and low growth rate groups (Figure 3D; representative histopathologic images shown in Figures 3E and F). Of the 20 pigs included in the study, 16 pigs (80%) had some degree of lymphoid depletion. Overall, mean lymphoid depletion scores indicated more severe pathology in the low growth rate group; 1.4 ± 0.7 and 1.2 ± 1.0 in low and high growth rate groups, respectively. However, no significant difference between the groups was detected (p = 0.58; Mann-Whitney U test).

Improved growth rates were associated with increased fecal microbiome diversity and shifts in microbial composition

Overall, the LLMDA identified 184 uniquely classified microbes across all 20 pigs. Identified bacterial species most commonly fell within the phyla Proteobacteria or Firmicutes. At the family level, fifty-nine unique classification groups were identified, with most being identified at the family level ( n = 52) and seven additional higher classifications which could not be further identified. From the fifty-nine classifications, most were bacterial {n = 52), but other represented groups included archaea (n = 1), eukaryotes (// = 3) and viruses (// = 3). Microbiome diversity was calculated using the LLMDA data as the number of families and species detected in the feces of each pig (Figures 4A and 4B). The mean number of families detected were similar in both growth rate groups; 29.2 ± 5.5 and 29.1 ± 2.6 in high and low growth rate groups, respectively (p = 0.96; Mann-Whitney U test). Similarly, no significant difference was detected in the mean number of species between the two groups (p = 0.73, Mann-Whitney U test). The Firmicutes/Bacteriodetes ratio was calculated by dividing the number of Firmicutes species by the number of Bacteroidetes species. High growth pigs had a higher numerical mean Firmicutes/Bacteroidetes ratio (2.6 ± 1.0 and 1.9 ± 0.7 in high and low growth rate groups, respectively), with a trend towards significance detected (p = 0.07; unpaired /-test; Figure 4C).

At the level of family microbial composition detected by the LLMDA, most microbial families were detected at similar prevalence rates among the two growth rate groups (Figure 5A). Several microbial families were detected in all 20 pigs, including Anaplasmataceae , unclassified Bacteria, B radyrh izo b iaceae , unclassified Clostridiales, Prevote llaceae, and Spirochaetaceae . Although not statistically significant, Mycoplasmataceae was detected in less than half of high growth rate pigs (4/10) compared to a greater detection rate in low growth rate pigs (8/10; p = 0.17, Fisher’s exact test). Additionally, Streptococcaceae was detected at a higher level in high growth rate pigs (7/10) compared to low growth pigs (3/10), albeit a lack of statistical significance (p = 0.18, Fisher’s exact test). Species diversity within each family was analyzed for differences associated with growth; two significant differences were detected (Figure 5B). First, high growth rate pigs had increased species diversity within the group of unclassified bacteria (p = 0.046; Mann-Whitney U test). Second, within the Mycop!asmataceae family, there was a trend towards less species diversity in high growth pigs; 0.5 ± 0.7 species in high growth pigs and 1.3 ± 0.8 species in low growth pigs (p = 0.059; Mann-Whitney U test). Increased species diversity was also noted in low growth pigs within the family Prevotellaceae , although no significant difference was detected (p = 0.12; Mann-Whitney U test).

At the species level on the LLMDA, three bacterial species were identified at greater rates in high growth pigs (data not shown). Bacteroides pectinophilus was detected in half of high growth rate pigs (5/10) while no low growth rate pigs had this species present (p = 0.03; Fisher’s exact test). Furthermore, two bacterial species in the family Lachnospiraceae trended towards having higher prevalence in high growth rate pigs. Lachnospiraceae bacterium C6A11 and Lachnospiraceae bacterium P6B14 were found in 6 and 7 of the high growth rate pigs, respectively, compared to 1 and 2 of the low growth rate pigs, respectively (p = 0.057 and p = 0.069, respectively; Fisher’s exact test). Although not significant, half the number of high growth pigs (4 versus 8 in low growth group) had Mycoplasma conjunctivae and two times the number of high growth pigs (8 versus 4 in low growth group) had Ruminococcaceae bacterium AE2021 detected in the feces (p = 0.169; Fisher’s exact test). These bacterial species may play a role in promoting or deterring growth performance after vaccination and co-challenge.

In addition to the LLMDA, fecal bacteriomes were further analyzed using 16S rDNA sequencing. Two of the samples from high growth rate pigs were excluded from the analysis due to failed quality control after 16S rDNA sequencing at UMGC. Therefore, 16S rDNA sequencing results were compared between 8 high growth rate pigs and 10 low growth rate pigs. Both bacterial and archaeal microbes were identified through 16S rDNA sequencing of fecal samples, with several bacterial families being detected across all 18 pigs, such as Clostridiaceae , Lachnospiraceae , Ruminococcaceae , Lactohacillaceae , Prevotellaceae , Streptococcaceae, and Veillonellaceae, Coriohacteriaceae , and Erysipelotrichaceae .

Based on the relative abundance of OTUs, the two most abundant phyla were Bacteroidetes and Firmicutes in the fecal samples from the 18 pigs; no significant differences in relative abundance of these two phyla were identified between the growth rate groups (adjusted p > 0.05; Likelihood Ratio Test; Figure 6A). However, the relative abundance of bacteria in the phylum Spirochaetes trended toward being significantly higher in the high growth group (adjusted p = 0.06; Likelihood Ratio Test). The relative abundance of Spirochaetes was 1.5% in high growth pigs while only 0.3% in low growth pigs. At the family level (Figure 6B), a lower relative abundance of Lachnospiraceae was detected in high growth rate pigs (5.8%) compared to low growth rate pigs (8.2%; adjusted p = 0.021; Likelihood Ratio Test).

Alpha diversity between the two groups was compared using Chaol and Shannon Index metrics (Figure 6C). The Chaol diversity metric based on OTU was significantly higher in high growth pigs compared to low growth pigs (p = 0.026; Mann-Whitney U test). Further, the observed OTU diversity was also significantly greater in high growth rate pigs (p = 0.023; Mann- Whitney U test; data not shown). This data suggests gut microbiome diversity may be beneficial for growth under the study conditions. Although the mean Shannon Index was numerically higher in high growth rate pigs, no significant difference was detected between the two groups (p = 0.145; Mann-Whitney U test). Beta diversity analysis did not detect any significant differences at the phyla or family level between the two groups (p > 0.05; PERMANOVA; data not shown).

Overall, microbial testing modalities utilized in the current study identified several gut microbiome characteristics associated with improved growth after vaccination and co-infection, including increased bacterial diversity, increased Barter oides pectinophilus, decreased Mycoplasmataceae species diversity, higher Firmicutes:Bacteroidetes ratios, increased relative abundance of the phylum Spirochaetes, reduced relative abundance of the family Lachnospiraceae, and increased Lachnospiraceae species C6A11 and P6B14.

Discussion

Despite PRRS MLV vaccines being widely used to reduce PRRS-associated losses in endemic herds, the currently available commercial vaccines are inadequate for disease control (Montaner-Tarbes et ah, Frontiers in veterinary science 6, 38, 2019) and additional tools are necessary to reduce the effects of PRRS on swine production. Through the gut-lung axis, a bi directional communication pathway between the gastrointestinal tract and pulmonary tissues, beneficial gut microbes provide an opportunity to improve immunity and efficacy of PRRS MLV vaccines. The present studies identified several microbiome characteristics in the post- vaccination/pre-challenge feces of pigs that had subsequent improvements in growth during the co-challenge period. These characteristics may predispose more rapid weight gain in the presence of wild-type PRRSV after vaccination. First, increased gut microbiome diversity was detected in high growth rate pigs. While increased gut microbiome diversity has shown to be beneficial after respiratory infection with diverse pathogens, such studies have not looked at the effects on infectious and noninfectious immunizations.

PRRSV is known to reduce nutrient digestibility and feed efficiency of growing pigs (Schweer et ah, Translational Animal Science 1, 480-488, 2017). In the present studies, the primary differentiator between high and low growth rate groups was indeed weight gain, as no significant differences were detected in humoral immunity, clinical disease or tonsillar pathology. As such, an interesting second microbiome characteristic associated with high growth included an increased Firmicutes to Bacteroidetes ratio.

A surprising finding of these studies was that high growth pigs had an increased relative abundance of bacteria in the phylum Spirochaetes , as this family is generally considered pathogenic (Hampson and Ahmed, Gut pathogens 1, 10, 2009). In contrast to the present findings, previous work reported decreased species diversity within the family Spirochateaceae in association with improved clinical disease outcome after fecal transplantation and PRRSV/PCV2 co-infection (Niederwerder et ah, Frontiers in microbiology 9, 1631, 2018). Moreover, in non-disease challenged conditions, high body weight pigs had significantly less Spirochaetes phylum abundance in feces compared to low body weight pigs (Oh et ah, Animal Science Journal = Nihon chikusan Gakkaiho 91, el3418, 2020).

Three bacterial species, including Bacteroides pectinophilus , Lachnospiraceae bacterium C6A11 and Lachnospiraceae bacterium P6B14, were detected at greater prevalence rates in high growth rate pigs. First, Bacteroides pectinophilus is an obligate anaerobe known to be an inhabitant of the human gut, which aids in the breakdown of pectin, an otherwise indigestible portion of plant cell walls. B. pectinophilus is considered unique from other Bacteroides species based on 16S rRNA genetic diversity and the presence of novel protein families and can diminish in gastrointestinal diseases such as irritable bowel syndrome. Lachnospiraceae bacterium C6A11 and Lachnospiraceae bacterium P6B 14 are unclassified anaerobic fermentative species in the Firmicutes phylum originally isolated from a cow rumen in New Zealand that play an important role in metabolism of plant material.

Described herein, several gut microbiome characteristics, such as increased diversity and shifts in microbial composition, were associated with a significant increase in ADG of pigs after PRRSV vaccination and PRRSV/PCV2 co-infection. These microbiome characteristics may contribute to improved outcome of pigs exposed to either attenuated vaccine or wild-type PRRS viruses.