G Model
VETMIC-6821; No. of Pages 8 Veterinary Microbiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Diversity and health status specific fluctuations of intrauterine microbial communities in postpartum dairy cows K. Wagener a,b, I. Prunner b, H. Pothmann b, M. Drillich b, M. Ehling-Schulz a,* a
Functional Microbiology, Institute of Microbiology, Department for Pathobiology, University of Veterinary Medicine Vienna, Veterina¨rplatz 1, 1210 Vienna, Austria b Clinical Unit for Herd Health Management in Ruminants, University Clinic for Ruminants, Department for Farm Animals and Veterinary Public Health, University of Veterinary Medicine Vienna, Veterina¨rplatz 1, 1210 Vienna, Austria
A R T I C L E I N F O
A B S T R A C T
Article history: Received 25 August 2014 Received in revised form 20 November 2014 Accepted 23 November 2014
For the interpretation of clinical findings of endometritis and the development of disease prevention and intervention strategies a better understanding of the dynamics and interactions within intrauterine bacterial communities in healthy and diseased cows is required. To gain deeper insights into fluctuations within the uterine microbiota, intrauterine samples were collected from 122 cows at the day of calving, days 3, 9, 15, 21 and 28 postpartum. A total of 2052 bacterial isolates were identified by Fourier-transforminfrared spectroscopy. This culturomics-based approach showed that the aerobic uterine microflora comprised a huge diversity of bacteria belonging to 202 different species, representing 76 genera, with members of the genus Staphylococcus (24.2%) being predominant. On species level the uterine microflora was dominated by Trueperella pyogenes (13.2%), Escherichia coli (11.2%), Staphylococcus xylosus (5.4%), Bacillus pumilus (5.2%) and Streptococcus uberis (4.9%). Comparative analysis of uterine bacteria from cows with different vaginal discharge scores (VDS) revealed health status specific temporal microbial diversifications. Although the intrauterine flora of all VDS groups was dominated by T. pyogenes, E. coli and Staphylococcus spp., the relative number of bacteria differed between VDS groups. The presence of T. pyogenes on days 15 and 21 significantly increased the risk of VDS 2 and 3 at day 21, whereas Staphylococci at day 9 reduced the likelihood of VDS 3 (P < 0.05). This study demonstrates that intrauterine bacterial infections are highly dynamic processes and that bacterial species follow specific patterns of progression, which require further research to decipher their potential role in uterine disease development. ß 2014 Elsevier B.V. All rights reserved.
Keywords: Intrauterine infection Uterine microbial diversity FTIR spectroscopy Microbial community dynamics
1. Introduction Bacterial uterine infections have a negative impact on uterine health and fertility in cows and consequently cause tremendous economic losses to dairy farmers. The uterus is
* Corresponding author. Tel.: +43 1 25077 2460; fax: +43 1 25077 2479. E-mail address: [email protected] (M. Ehling-Schulz).
supposed to be sterile before birth but during or immediately after birth it becomes rapidly colonized through the open cervix with a broad diversity of bacteria (Fo¨ldi et al., 2006). Within the first three weeks after calving 80–100% of the cows are positively tested for bacteria but not all animals show clinical symptoms of puerperal diseases such as metritis or clinical endometritis. Therefore, unspecific contamination and uterine infection must be distinguished (Fo¨ldi et al., 2006). The severity of clinical endometritis is classified by the quality
http://dx.doi.org/10.1016/j.vetmic.2014.11.017 0378-1135/ß 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: Wagener, K., et al., Diversity and health status specific fluctuations of intrauterine microbial communities in postpartum dairy cows. Vet. Microbiol. (2014), http://dx.doi.org/10.1016/j.vetmic.2014.11.017
G Model
VETMIC-6821; No. of Pages 8 2
K. Wagener et al. / Veterinary Microbiology xxx (2014) xxx–xxx
of vaginal discharge usually diagnosed from day 21 to 26 postpartum (LeBlanc et al., 2002; Williams et al., 2005; Sheldon et al., 2009; Leutert et al., 2012). Monitoring changes in the bacterial composition of the uterine fluid before the appearance of clinical signs of a disease is crucial not only for a critical interpretation of clinical findings but also for the development of optimal prevention and intervention strategies (Madoz et al., 2014). The natural defense mechanisms eliminate many of the microorganisms but the efficiency of elimination depends on the bacterial load and pathogenicity as well as on the immune status of the cow (Sheldon et al., 2002; LeBlanc et al., 2011). Escherichia coli, Trueperella pyogenes, Fusobacterium necrophorum and Prevotella spp. are well known uterine pathogens. Members of the genus Bacillus, Streptococcus, Enterococcus and coagulase negative staphylococci (CNS), which are among the most frequently isolated intrauterine bacteria, are discussed as potential pathogens or opportunistic contaminants (Westermann et al., 2010; Werner et al., 2012). The diversity of the latter bacterial groups and their dynamics in the early postpartum period is poorly understood. Although uterine infection is a highly complex process with contamination, elimination and re-contamination (Fo¨ldi et al., 2006), several studies have focused on the late puerperium (days 21 and 28) to identify bacteria and characterize uterine diseases (Williams et al., 2005; Westermann et al., 2010). Possible antagonistic interactions within the uterus have been reported for alpha hemolytic Streptococci and T. pyogenes. Preceding E. coli infections are assumed to facilitate subsequent persistent T. pyogenes infections (Bonnett et al., 1991; Dohmen et al., 2000). So far knowledge about the etiology of clinical endometritis derives primarily from studies focusing on known pathogens, such as E. coli and T. pyogenes (Sheldon et al., 2010; Amos et al., 2014), while detailed information on potential interactions between beneficial, opportunistic and pathogenic bacterial species is still lacking. Thus, we employed Fourier-transform infrared (FTIR) spectroscopy to explore the dynamics within the uterine microflora during the first four weeks postpartum. FTIR spectroscopy is a metabolome-based method that allows the identification and discrimination of bacteria down to species and subspecies level (Naumann et al., 1991; Ehling-Schulz et al., 2005; Wenning and Scherer, 2013) and it even can be used to identify hitherto unknown bacterial species and genera (Wagener et al., 2014a). Due to its high-throughput capacities and its high discriminatory power FTIR spectroscopy is also a suitable and effective tool for investigating complex bacterial communities (Fricker et al., 2011; Wagener et al., 2014b). It could be used to generate comprehensive collections of bacteria with known metabolic fingerprints for further downstream analyses and microbial interaction studies. The objective of this study was to monitor the fluctuation in the intrauterine bacterial community in the postpartum period by means of FTIR spectroscopy. To gain a deep insight into the bacterial community dynamics six sampling points during the first 28 days postpartum were defined and included in the analysis. Furthermore, the results from the microbial survey were evaluated in the context of vaginal discharge character of each cow to
decipher the impact of microbiota on clinical signs of endometritis. 2. Materials and methods The study was approved by the institutional ethics committee and the national authority according to § 8 of Law for Animal Experiments (Tierversuchsgesetz-TVG BMWF-68.205/246-II/3b/2010). 2.1. Study farm The study was conducted on the Teaching and Research Farm of the University of Veterinary Medicine Vienna between February 2011 and February 2013. The herd consisted of 70 lactating dairy cows (60% Simmental, 15% Holstein Friesian, and 25% Brown Swiss), housed in freestall barns with straw-bedded cubicles. The average milk yield was 8279 kg per lactation. The ration consisted of grass and maize silage, hay and was supplemented with minerals. Concentrates were fed by an automatic transponder station according to the individual milk yield. Cows were bred by artificial insemination and calved throughout the year in a straw bedded group maternity pen. 2.2. Study population and study design A total of 170 calvings were enrolled, including 46 (27.1%) primiparous cows. Cows with cesarean section, vaginal lacerations, downer cows and animals that received an antibiotic treatment were excluded. Intrauterine samples were taken within 12 h after calving (day 0), and on days 3, 9, 15, 21 and 28 after parturition. Prior to each sampling procedure, the rectal temperature was measured and a short general clinical examination was performed. Cows with >39.5 8C, fetid red-brown watery uterine discharge and systemic signs of illness were classified as having metritis according to Sheldon et al. (2006). The latter animals (n = 15) received a systemic antibiotic treatment with 1 mg/kg of Ceftiofur and were consequently excluded from further sampling. Other reasons for antibiotic treatment were mastitis (n = 10), orthopedic diseases (n = 7), and others (n = 10). From 6 cows no complete dataset was available because sampling was not possible. Thus, the final study population included in the analysis comprised 122 animals. 2.3. Gynecological examination On day 21 postpartum, a gynecological examination consisting of a vaginal inspection and rectal palpation of the uterus was performed. Vaginoscopic examination was carried out with a speculum and a torch as described by Westermann et al. (2010). Vaginal discharge was classified on a 4-point scale according to Williams et al. (2005). In brief, cows with clear discharge on day 21 were assigned to vaginal discharge score (VDS) 0, discharge with flecks of pus was defined as VDS 1 and animals with discharge containing 50% pus were classified as VDS 2 and 3.
Please cite this article in press as: Wagener, K., et al., Diversity and health status specific fluctuations of intrauterine microbial communities in postpartum dairy cows. Vet. Microbiol. (2014), http://dx.doi.org/10.1016/j.vetmic.2014.11.017
G Model
VETMIC-6821; No. of Pages 8 K. Wagener et al. / Veterinary Microbiology xxx (2014) xxx–xxx
2.4. Bacteriological samples Intrauterine samples for microbiological analysis were taken by using the previously described cytobrushtechnique (Kasimanickam et al., 2004; Westermann et al., 2010). Uterine sampling and laboratory steps were performed as described in detail by Prunner et al. (2014). In brief, cytobrushes (Gynobrush, Heinz Herenz, Hamburg, Germany; 20 mm in length and 7 mm in diameter) were screwed on a metal rod of 65 cm length. Prior to uterine sampling, the vulva of each cow was cleaned with a paper towel. To avoid contamination during sampling, the sampling device was protected by a disposable plastic catheter and a plastic sleeve. Inside the uterine cavity, the plastic sleeve was retracted and the brush was pushed forward and rolled onto the uterine wall. Then, the brush was retracted into the catheter and the device was removed from the reproductive tract. Immediately after sampling, the cytobrushes were transferred to a sterile PBS buffer solution in a room outside the stable. Within 1 h, material from the brush was rolled on sheep blood agarand MacConkey agar plates (Oxoid, Hampshire, United Kingdom) and incubated at 37 8C for 48 h. For preparation of bacterial cultures for subsequent FTIR analysis, sterile stainless steel Drigalski spatula were used. Identification of bacterial isolates by FTIR spectroscopy was carried out as previously described (Ehling-Schulz et al., 2005; Fricker et al., 2011) using a Tensor 27 FTIR spectrometer (Bruker Optics, Billerica, MA) coupled to a HTS-XT microplate adapter. Spectra from bacteria were recorded in the transmission mode covering the spectral range from 4000 to 500 cm1 and spectral data analysis was performed using the OPUS software (version 5.5, Bruker Optics, Billerica, MA) Reference spectral libraries were used for bacterial identification. In case that reference spectra from the databases did not reveal unequivocal results on species level, species identification was confirmed by 16S rRNA sequencing. Partial 16S rRNA was amplified using the universal 16S rRNA primers 8-26/56 (50 –AGA GTT TGA TCC TGG CTC A–30 ) and 1511-1493 (50 – CGG CTA CCT TGT TAC GAC–30 ) derived from E. coli (Stackebrandt and Liesack, 1992). DNA isolation and PCR reactions were carried out as described by Oberreuter et al. (2002). Subsequent sequencing, sequence editing and similarity searches were performed as described by Wagener et al. (2014a). 2.5. Statistical analysis For statistical analysis SPSS for Windows (version 21.0, SPSS Inc.) was used. Associations between infections with Staphylococcus spp., T. pyogenes, E. coli, Bacillus spp., Streptococcus spp., and Corynebacterium spp. were tested by x2-test. Multinomial logistic regression was used to assess associations between the presence of Staphylococcus spp., T. pyogenes, E. coli, Bacillus spp., Streptococcus spp., and Corynebacterium spp. at days 0, 3, 9, 15, and 21 postpartum (0 = negative; 1 = positive) and VDS. Variables were stepwise removed from the models in descending order (backward elimination) according to the results of likelihood ratio test. Remaining results with P < 0.05 for VDS 1,
3
2, 3 are reported as relative risk ratios (RRR) with VDS 0 as reference category and 95% confidence intervals (CI). The level of significance was set at P < 0.05. 3. Results 3.1. Gynecological examination At the gynecological examination at day 21 postpartum, 53 (43.4%) cows were diagnosed with clear vaginal discharge (VDS 0), 24 (19.7%) cows were classified as VDS 1, 23 (18.9%) and 22 (18.0%) animals showed VDS 2 and VDS 3. 3.2. Diversity of the intrauterine bacterial community Uterine sampling of 122 cows at six sampling days (732 samples) resulted in a total number of 2052 isolates grown under aerobic culture conditions, representing 202 species and 76 genera. Fig. 1 shows the 22 most frequently detected genera with prevalences >0.5%. The most predominant microorganisms were bacteria belonging to the genus Staphylococcus (n = 496; 24.2%), followed by Trueperella spp. (n = 270; 13.2%) and Escherichia spp. (n = 230; 11.2%). Other frequently isolated genera were Bacillus spp. (n = 191; 9.3%), Streptococcus spp. (n = 162; 7.9%), and Corynebacterium spp. (n = 161; 7.8%). A total of 56 genera with prevalences 0.5% (22 out of 76) are presented individual, whereas genera with prevalence
VETMIC-6821; No. of Pages 8 Veterinary Microbiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Diversity and health status specific fluctuations of intrauterine microbial communities in postpartum dairy cows K. Wagener a,b, I. Prunner b, H. Pothmann b, M. Drillich b, M. Ehling-Schulz a,* a
Functional Microbiology, Institute of Microbiology, Department for Pathobiology, University of Veterinary Medicine Vienna, Veterina¨rplatz 1, 1210 Vienna, Austria b Clinical Unit for Herd Health Management in Ruminants, University Clinic for Ruminants, Department for Farm Animals and Veterinary Public Health, University of Veterinary Medicine Vienna, Veterina¨rplatz 1, 1210 Vienna, Austria
A R T I C L E I N F O
A B S T R A C T
Article history: Received 25 August 2014 Received in revised form 20 November 2014 Accepted 23 November 2014
For the interpretation of clinical findings of endometritis and the development of disease prevention and intervention strategies a better understanding of the dynamics and interactions within intrauterine bacterial communities in healthy and diseased cows is required. To gain deeper insights into fluctuations within the uterine microbiota, intrauterine samples were collected from 122 cows at the day of calving, days 3, 9, 15, 21 and 28 postpartum. A total of 2052 bacterial isolates were identified by Fourier-transforminfrared spectroscopy. This culturomics-based approach showed that the aerobic uterine microflora comprised a huge diversity of bacteria belonging to 202 different species, representing 76 genera, with members of the genus Staphylococcus (24.2%) being predominant. On species level the uterine microflora was dominated by Trueperella pyogenes (13.2%), Escherichia coli (11.2%), Staphylococcus xylosus (5.4%), Bacillus pumilus (5.2%) and Streptococcus uberis (4.9%). Comparative analysis of uterine bacteria from cows with different vaginal discharge scores (VDS) revealed health status specific temporal microbial diversifications. Although the intrauterine flora of all VDS groups was dominated by T. pyogenes, E. coli and Staphylococcus spp., the relative number of bacteria differed between VDS groups. The presence of T. pyogenes on days 15 and 21 significantly increased the risk of VDS 2 and 3 at day 21, whereas Staphylococci at day 9 reduced the likelihood of VDS 3 (P < 0.05). This study demonstrates that intrauterine bacterial infections are highly dynamic processes and that bacterial species follow specific patterns of progression, which require further research to decipher their potential role in uterine disease development. ß 2014 Elsevier B.V. All rights reserved.
Keywords: Intrauterine infection Uterine microbial diversity FTIR spectroscopy Microbial community dynamics
1. Introduction Bacterial uterine infections have a negative impact on uterine health and fertility in cows and consequently cause tremendous economic losses to dairy farmers. The uterus is
* Corresponding author. Tel.: +43 1 25077 2460; fax: +43 1 25077 2479. E-mail address: [email protected] (M. Ehling-Schulz).
supposed to be sterile before birth but during or immediately after birth it becomes rapidly colonized through the open cervix with a broad diversity of bacteria (Fo¨ldi et al., 2006). Within the first three weeks after calving 80–100% of the cows are positively tested for bacteria but not all animals show clinical symptoms of puerperal diseases such as metritis or clinical endometritis. Therefore, unspecific contamination and uterine infection must be distinguished (Fo¨ldi et al., 2006). The severity of clinical endometritis is classified by the quality
http://dx.doi.org/10.1016/j.vetmic.2014.11.017 0378-1135/ß 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: Wagener, K., et al., Diversity and health status specific fluctuations of intrauterine microbial communities in postpartum dairy cows. Vet. Microbiol. (2014), http://dx.doi.org/10.1016/j.vetmic.2014.11.017
G Model
VETMIC-6821; No. of Pages 8 2
K. Wagener et al. / Veterinary Microbiology xxx (2014) xxx–xxx
of vaginal discharge usually diagnosed from day 21 to 26 postpartum (LeBlanc et al., 2002; Williams et al., 2005; Sheldon et al., 2009; Leutert et al., 2012). Monitoring changes in the bacterial composition of the uterine fluid before the appearance of clinical signs of a disease is crucial not only for a critical interpretation of clinical findings but also for the development of optimal prevention and intervention strategies (Madoz et al., 2014). The natural defense mechanisms eliminate many of the microorganisms but the efficiency of elimination depends on the bacterial load and pathogenicity as well as on the immune status of the cow (Sheldon et al., 2002; LeBlanc et al., 2011). Escherichia coli, Trueperella pyogenes, Fusobacterium necrophorum and Prevotella spp. are well known uterine pathogens. Members of the genus Bacillus, Streptococcus, Enterococcus and coagulase negative staphylococci (CNS), which are among the most frequently isolated intrauterine bacteria, are discussed as potential pathogens or opportunistic contaminants (Westermann et al., 2010; Werner et al., 2012). The diversity of the latter bacterial groups and their dynamics in the early postpartum period is poorly understood. Although uterine infection is a highly complex process with contamination, elimination and re-contamination (Fo¨ldi et al., 2006), several studies have focused on the late puerperium (days 21 and 28) to identify bacteria and characterize uterine diseases (Williams et al., 2005; Westermann et al., 2010). Possible antagonistic interactions within the uterus have been reported for alpha hemolytic Streptococci and T. pyogenes. Preceding E. coli infections are assumed to facilitate subsequent persistent T. pyogenes infections (Bonnett et al., 1991; Dohmen et al., 2000). So far knowledge about the etiology of clinical endometritis derives primarily from studies focusing on known pathogens, such as E. coli and T. pyogenes (Sheldon et al., 2010; Amos et al., 2014), while detailed information on potential interactions between beneficial, opportunistic and pathogenic bacterial species is still lacking. Thus, we employed Fourier-transform infrared (FTIR) spectroscopy to explore the dynamics within the uterine microflora during the first four weeks postpartum. FTIR spectroscopy is a metabolome-based method that allows the identification and discrimination of bacteria down to species and subspecies level (Naumann et al., 1991; Ehling-Schulz et al., 2005; Wenning and Scherer, 2013) and it even can be used to identify hitherto unknown bacterial species and genera (Wagener et al., 2014a). Due to its high-throughput capacities and its high discriminatory power FTIR spectroscopy is also a suitable and effective tool for investigating complex bacterial communities (Fricker et al., 2011; Wagener et al., 2014b). It could be used to generate comprehensive collections of bacteria with known metabolic fingerprints for further downstream analyses and microbial interaction studies. The objective of this study was to monitor the fluctuation in the intrauterine bacterial community in the postpartum period by means of FTIR spectroscopy. To gain a deep insight into the bacterial community dynamics six sampling points during the first 28 days postpartum were defined and included in the analysis. Furthermore, the results from the microbial survey were evaluated in the context of vaginal discharge character of each cow to
decipher the impact of microbiota on clinical signs of endometritis. 2. Materials and methods The study was approved by the institutional ethics committee and the national authority according to § 8 of Law for Animal Experiments (Tierversuchsgesetz-TVG BMWF-68.205/246-II/3b/2010). 2.1. Study farm The study was conducted on the Teaching and Research Farm of the University of Veterinary Medicine Vienna between February 2011 and February 2013. The herd consisted of 70 lactating dairy cows (60% Simmental, 15% Holstein Friesian, and 25% Brown Swiss), housed in freestall barns with straw-bedded cubicles. The average milk yield was 8279 kg per lactation. The ration consisted of grass and maize silage, hay and was supplemented with minerals. Concentrates were fed by an automatic transponder station according to the individual milk yield. Cows were bred by artificial insemination and calved throughout the year in a straw bedded group maternity pen. 2.2. Study population and study design A total of 170 calvings were enrolled, including 46 (27.1%) primiparous cows. Cows with cesarean section, vaginal lacerations, downer cows and animals that received an antibiotic treatment were excluded. Intrauterine samples were taken within 12 h after calving (day 0), and on days 3, 9, 15, 21 and 28 after parturition. Prior to each sampling procedure, the rectal temperature was measured and a short general clinical examination was performed. Cows with >39.5 8C, fetid red-brown watery uterine discharge and systemic signs of illness were classified as having metritis according to Sheldon et al. (2006). The latter animals (n = 15) received a systemic antibiotic treatment with 1 mg/kg of Ceftiofur and were consequently excluded from further sampling. Other reasons for antibiotic treatment were mastitis (n = 10), orthopedic diseases (n = 7), and others (n = 10). From 6 cows no complete dataset was available because sampling was not possible. Thus, the final study population included in the analysis comprised 122 animals. 2.3. Gynecological examination On day 21 postpartum, a gynecological examination consisting of a vaginal inspection and rectal palpation of the uterus was performed. Vaginoscopic examination was carried out with a speculum and a torch as described by Westermann et al. (2010). Vaginal discharge was classified on a 4-point scale according to Williams et al. (2005). In brief, cows with clear discharge on day 21 were assigned to vaginal discharge score (VDS) 0, discharge with flecks of pus was defined as VDS 1 and animals with discharge containing 50% pus were classified as VDS 2 and 3.
Please cite this article in press as: Wagener, K., et al., Diversity and health status specific fluctuations of intrauterine microbial communities in postpartum dairy cows. Vet. Microbiol. (2014), http://dx.doi.org/10.1016/j.vetmic.2014.11.017
G Model
VETMIC-6821; No. of Pages 8 K. Wagener et al. / Veterinary Microbiology xxx (2014) xxx–xxx
2.4. Bacteriological samples Intrauterine samples for microbiological analysis were taken by using the previously described cytobrushtechnique (Kasimanickam et al., 2004; Westermann et al., 2010). Uterine sampling and laboratory steps were performed as described in detail by Prunner et al. (2014). In brief, cytobrushes (Gynobrush, Heinz Herenz, Hamburg, Germany; 20 mm in length and 7 mm in diameter) were screwed on a metal rod of 65 cm length. Prior to uterine sampling, the vulva of each cow was cleaned with a paper towel. To avoid contamination during sampling, the sampling device was protected by a disposable plastic catheter and a plastic sleeve. Inside the uterine cavity, the plastic sleeve was retracted and the brush was pushed forward and rolled onto the uterine wall. Then, the brush was retracted into the catheter and the device was removed from the reproductive tract. Immediately after sampling, the cytobrushes were transferred to a sterile PBS buffer solution in a room outside the stable. Within 1 h, material from the brush was rolled on sheep blood agarand MacConkey agar plates (Oxoid, Hampshire, United Kingdom) and incubated at 37 8C for 48 h. For preparation of bacterial cultures for subsequent FTIR analysis, sterile stainless steel Drigalski spatula were used. Identification of bacterial isolates by FTIR spectroscopy was carried out as previously described (Ehling-Schulz et al., 2005; Fricker et al., 2011) using a Tensor 27 FTIR spectrometer (Bruker Optics, Billerica, MA) coupled to a HTS-XT microplate adapter. Spectra from bacteria were recorded in the transmission mode covering the spectral range from 4000 to 500 cm1 and spectral data analysis was performed using the OPUS software (version 5.5, Bruker Optics, Billerica, MA) Reference spectral libraries were used for bacterial identification. In case that reference spectra from the databases did not reveal unequivocal results on species level, species identification was confirmed by 16S rRNA sequencing. Partial 16S rRNA was amplified using the universal 16S rRNA primers 8-26/56 (50 –AGA GTT TGA TCC TGG CTC A–30 ) and 1511-1493 (50 – CGG CTA CCT TGT TAC GAC–30 ) derived from E. coli (Stackebrandt and Liesack, 1992). DNA isolation and PCR reactions were carried out as described by Oberreuter et al. (2002). Subsequent sequencing, sequence editing and similarity searches were performed as described by Wagener et al. (2014a). 2.5. Statistical analysis For statistical analysis SPSS for Windows (version 21.0, SPSS Inc.) was used. Associations between infections with Staphylococcus spp., T. pyogenes, E. coli, Bacillus spp., Streptococcus spp., and Corynebacterium spp. were tested by x2-test. Multinomial logistic regression was used to assess associations between the presence of Staphylococcus spp., T. pyogenes, E. coli, Bacillus spp., Streptococcus spp., and Corynebacterium spp. at days 0, 3, 9, 15, and 21 postpartum (0 = negative; 1 = positive) and VDS. Variables were stepwise removed from the models in descending order (backward elimination) according to the results of likelihood ratio test. Remaining results with P < 0.05 for VDS 1,
3
2, 3 are reported as relative risk ratios (RRR) with VDS 0 as reference category and 95% confidence intervals (CI). The level of significance was set at P < 0.05. 3. Results 3.1. Gynecological examination At the gynecological examination at day 21 postpartum, 53 (43.4%) cows were diagnosed with clear vaginal discharge (VDS 0), 24 (19.7%) cows were classified as VDS 1, 23 (18.9%) and 22 (18.0%) animals showed VDS 2 and VDS 3. 3.2. Diversity of the intrauterine bacterial community Uterine sampling of 122 cows at six sampling days (732 samples) resulted in a total number of 2052 isolates grown under aerobic culture conditions, representing 202 species and 76 genera. Fig. 1 shows the 22 most frequently detected genera with prevalences >0.5%. The most predominant microorganisms were bacteria belonging to the genus Staphylococcus (n = 496; 24.2%), followed by Trueperella spp. (n = 270; 13.2%) and Escherichia spp. (n = 230; 11.2%). Other frequently isolated genera were Bacillus spp. (n = 191; 9.3%), Streptococcus spp. (n = 162; 7.9%), and Corynebacterium spp. (n = 161; 7.8%). A total of 56 genera with prevalences 0.5% (22 out of 76) are presented individual, whereas genera with prevalence
