https://orcid.org/0000-0002-4993-0511
Figures
Abstract
Sepsis, a dysregulated host response to infection, severely affects calf health. To date, sepsis diagnosis relies on clinical examination and positive blood culture. Differently, in humans 16S rRNA gene analysis is a valuable adjunct to blood culture as it allows for broader assessment of bacterial DNA in whole blood and its fractions. However, its efficacy in cattle remains unknown. Therefore, this study characterized and compared the bacterial DNA detected in whole blood and its fractions between healthy calves and those showing clinical signs of sepsis. The study sample was 18 Piedmontese calves classified according to their clinical status as suspected septic (S, 8/18) or healthy (H, 10/18). Aseptic blood samples were collected into EDTA tubes for 16S rRNA gene analysis of whole blood (WB), plasma (PL), buffy coat (BC), and red blood cells (RBC). Aseptic samples were additionally taken only from the S calves for blood culture. Clinical and microbiological parameters were compared between the two groups and between the blood fractions within each group. The S calves were diagnosed with pneumonia (3/8, 37.5%), enteritis (3/8, 37.5%), and omphalitis (2/8, 25%). Microbiome analysis revealed significant intra-group differences in α and β diversity indices between PL and the other blood fractions for both groups. Comparison between the S and the H calves showed differences in β diversity indices for PL and WB. DNA of known pathogens (e.g., Escherichia coli) and species not commonly associated with sepsis (e.g., Cutibacterium acnes) were more abundant in the S calves. Moreover, in S calves, 16S rRNA gene sequencing detected E. coli DNA more often (8/8, 100%) than blood culture (2/8, 25%). While the DNA of several bacteria can be detected in calves showing clinical signs of sepsis, further studies are needed to clarify its origin, role, and distribution in blood fractions.
Citation: Borriello G, Valentini F, Ferrini S, Di Muro G, Cagnotti G, Grego E, et al. (2025) Characterization of blood microbial population in beef calves with clinical signs of sepsis using 16S rRNA gene sequencing. PLoS One 20(5): e0324469. https://doi.org/10.1371/journal.pone.0324469
Editor: Ulrike Gertrud Munderloh, University of Minnesota, UNITED STATES OF AMERICA
Received: September 14, 2024; Accepted: April 24, 2025; Published: May 23, 2025
Copyright: © 2025 Borriello et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Calves are particularly susceptible to infection during their first weeks of life. In severe cases, a dysregulated host response can lead to life-threatening organ dysfunction, also known as sepsis [1]. Despite its impact on animal health, sepsis remains under-investigated in farm animals. Clinical diagnosis relies on the detection of severe clinical alterations along with bacterial assessment in the bloodstream by blood culture [1,2]. Such methods have limitations, however. Indeed, clinical parameters (e.g., respiratory and heart rates) can be influenced by the animal’s behavioral response to clinical examination, leading to potentially inaccurate assessment [3]. The scoring systems used in human medicine are more accurate but have not yet been validated in cattle [4]. Blood culture, although it is the gold standard, may have suboptimal sensitivity and specificity, especially in cases of slow-growing bacteria or low bacterial load [5,6]. Moreover, the risk of blood culture contamination seems to be higher for animals than for humans [7,8]. However, it remains unclear whether the higher risk is due to the environment, differences in the microbial skin population or the presence of hair surrounding the shaved area. In recent years, 16S rRNA gene sequencing has gained attention because of its ability to partially overcome the limitations of blood culture. Human studies have reported that it can detect traces of bacterial DNA in healthy and diseased subjects [9–11] and identify bacterial DNA in the blood of septic patients, even when blood cultures are negative [12]. Further, 16S rRNA gene sequencing has proved reliable for comprehensive analysis of the entire bacterial DNA within a sample and highlighted differences between the bacterial DNA found in blood of healthy and septic human patients [13]. In addition, it revealed differences in the microbiome of blood fractions in physiological and pathological conditions, even though the underlying reasons for these differences still need to be clarified [9,14,15]. In bovine species, 16S rRNA gene sequencing has been applied to characterize the microbiome of body districts in healthy and diseased animals [10,16–18]. Nevertheless, to our best knowledge, there are no studies on blood bacterial populations in septic calves. Understanding the characterization of bacterial populations by 16S rRNA gene sequencing could be the first step to comprehend the role that microorganisms play in sepsis. For these reasons, with the present study, we aimed to characterize and compare the bacterial DNA detected in whole blood and its fractions between healthy calves and those suspected septic.
Materials and methods
Study population
This analytical cross-sectional study was conducted from December 2022 to July 2023 and involved Piedmontese breed calves reared in northern Italy (Piedmont) referred for suspected sepsis or examined during routine herd visits. Calves were considered eligible if they were within the neonatal period (3 to 30 days old) [19] and had not been receiving antimicrobial treatment before or at the time of examination. All the procedures were conducted on the farms or at the Veterinary Teaching Hospital, University of Turin, in conformity with good clinical practices [20]; the study was approved by the Ethical Animal Care and Use Committee, University of Turin (Prot. n. 0004177). The owners were informed about the procedures, the potential risks of the study and signed an informed consent form for veterinary assessment of their animals.
Clinical examination and hematobiochemical analysis
Medical history of each calf was collected on enrollment into the study. A complete clinical examination and blood sampling for hematobiochemical analysis to determine the clinical status were performed by the same operator (first author, GB). Blood was collected from the jugular vein with a 21G needle and Vacutainer® (Becton and Dickinson, Franklin Lakes, NJ, USA) EDTA and serum tubes. The samples were analyzed at the Internal Medicine Unit, Department of Veterinary Sciences of Turin laboratory on an ADVIA® 120 (Siemens, Munich, Germany) for biochemical analysis and, for complete blood count (CBC) an BT3500Vet (Futurlab, Limena, Italy).
The calves were classified as critically ill if severe clinical alterations and at least one focal site of infection were found. When needed, further blood tests or other diagnostic techniques (i.e., ultrasonography, radiography) were performed, using appropriate anesthetic and analgesic protocols if needed, to identify the infectious focus in the CI calves. Then, each CI calf was classified based on the systemic inflammatory response syndrome (SIRS) score [21]. If a calf was classified as affected by the SIRS, the probability of bacteremia was estimated using the score described in Fecteau [22] (Table 1). Animals were classified as suspected septic (S) if the probability score was higher than 40.8%, as specified by the original author [22].
Conversely, calves with no clinical signs were enrolled in the healthy (H) control group. Whenever possible, the H calves were selected from the same farm as the S calves, matched for age and sex. If this was not feasible, H calves were enrolled from nearby farms where an S calf was involved, ensuring that the managerial procedures and the environmental conditions were similar.
Sterile blood sampling
Sterile blood samples were collected from the S and the H calves. For each calf, the hairs from the jugular groove were completely removed using an electronic shaver and a handled razor. The skin was then scrubbed as described by Garcia [23] to ensure asepsis of the sampling site. Blood samples were collected with a 21G needle into two Vacutainer® EDTA tubes for 16S rRNA gene analysis. The samples were refrigerated at 4° C during transport. An additional aseptic sampling for blood culture was performed only for the S calves. To minimize the risk of contamination, the scrub was repeated and blood was drawn using a 21G needle and a 10-ml syringe. The blood was immediately added to a culture bottle containing enriched media (Oxoid Signal® Blood Culture System. Oxoid Limited, Basingstoke, UK). All the samples were delivered to the laboratories of the Infectious Disease Unit, Department of Veterinary Sciences of Turin within 1 h of collection.
Blood culture
The samples were processed upon arrival at the laboratory. Briefly, each bottle underwent initial incubation at 36 ± 1° C for 1 h, followed by incubation with agitation (150 rpm) for 24 h. Then, the bottles were incubated without agitation for six days with daily observation. The presence of turbidity in the medium and/or fluid in the upper cylinder, indicates microbial activity. In such cases, subcultures were performed on blood agar (Oxoid, Basingstoke, UK) under both standard aerobic and anaerobic conditions. Identification of all bacterial colonies was performed using MALDI-TOF mass spectrometry (Bruker Daltonik GmbH, Bremen, Germany).
16S rRNA gene sequencing and data analysis
Blood fractions preparation.
From each calf, the first EDTA tube collected for 16S rRNA gene sequencing was divided into 500-µl aliquots for the whole blood fraction (WB). The second tube was centrifuged (3000 g x 15 min at 4° C) to separate the whole blood into fractions. Plasma (PL) was collected and centrifuged (8000 g x 5 min) at room temperature to pellet the remaining cells; the supernatant was used for analysis. The buffy coat (BC) phase was collected after removing the remaining plasma, and the red blood cell (RBC) fraction was obtained by pipetting from the bottom of the centrifuged tube. All samples were stored in DNA-RNA free Eppendorf and frozen at -80° C.
DNA extraction.
DNA was extracted from WB, BC, RBC, and PL fractions using a DNeasy Blood & Tissue Kit (Qiagen Ltd., Manchester, UK), according to the manufacturer’s instructions. DNA samples were eluted in 50 µL and stored at −20° C until further analysis. The DNA concentration of each sample was quantified using a Qubit 2.0 Fluorometer (ThermoFisher Scientific, Waltham, MA, USA) with the Qubit dsDNA HS Assay Kit (ThermoFisher Scientific). DNA integrity was evaluated by agarose gel electrophoresis (1% agarose in tris buffer EDTA 0.5X); DNA purity was determined by measuring the ratio of absorbance at 260/280 using a NanoDrop spectrophotometer (ThermoFisher Scientific). The samples meeting quality criteria were submitted for library preparation and metagenomic 16S rRNA gene sequencing.
Library preparation and sequencing.
DNA library preparation, 16S rRNA gene sequencing, and quality control were performed by Macrogen Europe (https://www.macrogen-europe.com/), a genomics service company. The library was prepared using Herculase II Fusion DNA Polymerase Nextera XT Index Kit V2 (Illumina, San Diego, CA, USA) following the 16S Metagenomic Sequencing Library Preparation Part # 15044223 Rev. B protocol. PCR amplification was performed to amplify the V3-V4 region of the 16S rRNA gene using the primers Bakt_341F: CCTACGGGNGGCWGCAG and Bakt_805R: GACTACHVGGGTATCTAATCC. The samples were run on an Illumina MiSeq platform to generate paired-end reads. The 16S rRNA gene sequencing raw data from Illumina sequencing comprised forward and reverse sequencing files. The sequences with high quality (quality score >20) were deemed suitable for further downstream analysis.
Bioinformatic analysis.
The fastq files were processed using QIIME2 (Quantitative Insights Into Microbial Ecology, release 2020.8) [24]. The DADA2 algorithm [25] (via q2-dada2 implementation) was used to filter the quality and to denoise the sequence data. In particular the forward and reverse reads were further cleaned by trimming off the first ten positions and truncating at positions 280 and 220, respectively, to remove low-quality portions.
The resulting amplicon sequence variants (ASVs) were cleaned from the chimeric sequences and taxonomically characterized by NCBI BLAST (Basic Local Alignment Search Tool, release 2.11.0) by aligning them on the NCBI 16S rRNA gene bacterial database (2023_09_30).
For each ASV, the best 20 alignments were selected and then filtered based on high levels of similarity (>= 99% identical matches, >= 99% query coverage, and eValue <0.05). This allowed one or more species to be associated with the ASVs characterized by at least one BLASTN result above the various cutoffs. Translation of GeneBank identifiers with Hugo Name is based on GeneGank db (2023_10_03). The abundance signal was further collapsed by summing the contributions of different ASVs attributed to the same species within each subject. α diversities were evaluated using the Chao1 Index, Shannon Index, and Inverse Simpson Index measures. Depending on the number of classes, their significance was analyzed using the Wilcoxon rank sum exact test or the Kruskal-Wallis test and subsequent pairwise comparison with the Wilcoxon rank sum exact test and Benjamini-Hockberg correction, when necessary.
β diversity was calculated with Bray-Curtis dissimilarity; its significance was analyzed using Permutational multivariate analysis of variance (PERMANOVA) based on 1000 permutations.
To mitigate the impact of contaminants and identify the bacteria linked to disease, the relative abundance of species in the samples from the S and the H subjects was compared. We utilized the modified Wilcoxon rank test (implemented through the ‘ziw’ function in the R package ZIR), specifically designed for Zero-Inflated Data (developed by Wanjie Wang, Eric Z. Chen, and Hongzhe Li in 2021). The resulting p-values were adjusted using the false discovery rate method (Benjamin-Hockberg correction). Statistical significance was set at alpha equal to 0.05, focusing exclusively on bacterial species that exhibited a significant increase in abundance in the samples associated to S calves. The final relative abundances of the significant species in each disease sample were calculated as follows: the median relative abundance observed in the H samples for each species was subtracted to each S sample, the resulting relative abundance for each species was recalculated to achieve a sum value of 1 in each disease sample.
Statistical analysis
Statistical analysis was performed in R (release 4.3.2) and RStudio (release 2023.06.1 Build 524) environments. Normal distribution of data was assessed using the Shapiro-Wilk test. Data are expressed as mean and standard deviation (SD) or median and interquartile range (IQR) for continuous variables according to their distribution, and as percentages and frequencies for categorical nominal variables. The R packages used were: phyloseq (version 1.44.0), microbiome (version 1.22.0), vegan (version 2.6-4) and chvlyl/ZIR (version 1.0.0). Alpha was set at 0.05.
Results
Animals
The study sample was 18 Piedmontese calves, 8 of which were classified as S and 10 as H. Calves were mostly male (S: 7/8, 87.5%; H: 6/10, 60%) and no differences were found in the mean age between the groups (S: 10.75 ± 3.09 days; H: 11.10 ± 5.38 days).
Clinical examination and blood culture
The S group calves were diagnosed with pneumonia (3/8, 37.5%), enteritis (3/8, 37.5%), and omphalitis (2/8, 25%). They had a probability of being bacteremic higher than 75%, based on the Fecteau score [22]. Blood cultures tested negative in 4/8 (50%); Escherichia coli (2/8, 25%), Streptococcus uberis (1/8, 12.5%), and Providencia stuartii (1/8, 12.5%) were isolated from the positive samples.
DNA extraction and quantification
For each animal, samples of the four matrices (WB, PL, BC, RBC) were sent for 16S rRNA gene analysis, therefore 72 samples were sent initially. The DNA was found suitable for sequencing according to the Macrogen Europe quality check in most of the samples (59/72, 82%): 35 samples were from the H group calves and 24 from the S group calves. The samples from the H group were mostly WB (10/35, 28.6%) and PL (10/35, 28.6%) fractions, while slightly fewer BC (8/35, 22.8%) and RBC (7/35, 20%) fractions. The samples from the S group were PL (8/24, 33.33%) and WB (7/24, 29.2), and BC fractions (6/24, 25%). Furthermore, only a few RBC fractions (3/24, 12.5%) could be submitted for analysis.
As shown in Table 2, for the S group, RBC had the maximum median value for the reads (2.87 x 105, IQR: 2.46 x 105 - 3.13 x 105) and WB had the minimum median for the reads (1.66 x 105, IQR: 1.48 x 105–1.80 x105). The BC values were significantly lower than those of the RBC (p < 0.05) or the PL (p ≤ 0.01); the WB fraction was lower than the PL fraction (p ≤ 0.01).
For the H group, the PL fraction had the maximum median for reads (2.97 x 105, IQR: 2.96 x 105–3.15 x 105) and the WB fraction had the minimum median (1.65 x 105, IQR 1.58 x 105–1.84 x 105). Significantly higher read values were noted for PL than for the other fractions (p ≤ 0.01). No significant differences were found when the same fractions were compared between the S and the H group.
Microbiome analysis
The 15 most abundant microorganisms detected in each blood fraction are shown in Fig 1. For the H group, the most abundant bacterial DNA was Phocaeicola vulgatus detected in WB, BC, and RBC, while Staphylococcus epidermidis was most abundant in PL. For the S group, the most abundant bacterial DNA was Escherichia coli detected in WB, Staphylococcus coagulans in PL, Bifidobacterium faecale in BC, and Flavobacterium gawaongense in RBC. The α and β diversity indices differed across blood fractions within the same group of calves. In particular, the Chao1 (p ≤ 0.01) and the Shannon indices (p < 0.05) were significantly higher in the PL fraction compared to the other fractions from the H group. In the S group, the Chao1 was significantly higher in PL than in RBC (p < 0.05), BC or WB (p ≤ 0.01). β diversity analysis revealed statistically distinct microbial compositions in PL compared to the other fractions (Fig 2).
S: suspected septic calves; H: healthy calves; WB: whole blood; PL: plasma; BC: buffy coat; RBC: red blood cells.
S: denotes suspected septic calves; H: healthy calves; WB: whole blood; PL: plasma; BC: buffy coat; RBC: red blood cells; (a,b: p ≤ 0.01; c,d: p < 0.05).
Comparison of the same blood fraction revealed no differences in the indices of α diversity between the S and the H group across all blood fractions, whereas significant differences between both groups were noted for the PL (p ≤ 0.01) and the WB (p < 0.05) fractions regarding β diversity analysis based on Bray-Curtis dissimilarity.
Differential abundance analysis
Differential abundance analysis revealed the presence of several bacterial species in the S group (Fig 3). The median relative abundance of Cutibacterium acnes DNA and Phascolarctobacterium succinatutens DNA were particularly high in WB. Considering the values of each calf, the DNA more frequently detected was from Cutibacterium acnes and Pseudomonas aeruginosa (Fig 4).
Escherichia coli DNA, S. coagulans DNA, and Proteus mirabilis DNA were particularly abundant in the PL fraction according to their median values and their abundance in each calf. Streptococcus ilei and Fusobacterium mortiferum DNA were more abundant in the BC. There was a variety of bacterial DNA in the RBC fraction, though their median relative abundance was generally low (Fig 3).
Comparison between blood culture and differential abundance analysis
The bacterial species isolated from blood culture were E.coli, S. uberis, and P. stuartii. In the two S calves (S_2, S_4) in which E. coli was detected at blood culture, its bacterial DNA was significantly more abundant in the PL fraction than in that of the H calves (Fig 4). E. coli DNA was not significantly more abundant in the other blood fractions from either S_2 or S_4. Regarding S_2 specifically, P. succinatutens and S. salivarius DNA were significantly more abundant in the WB fraction, whereas Ligilactobacillus ruminis DNA and Flavobacterium gawalongense DNA were the most detected in RBC. For S_4, the most abundant bacterial DNA were P. aeruginosa and C. acnes in WB, while Pseudomonas argentiniensis and Hydrobacter penzbergensis DNA were most often detected in BC (Fig 4).
S. uberis was detected in S_6 at blood culture but not significantly more abundant than that in any blood fractions of the H group. The most abundant bacterial DNA detected in S_6 was S. pasteuri and Roseburia intestinalis in WB, whereas E. coli and S. coagulans DNA was most often detected in PL (Fig 4).
P. stuartii was detected in S_7 at blood culture but its DNA was not significantly more abundant compared to H group. Most often detected in the WB of this calf was C. acnes and P. succinatutens DNA, whereas Myroides odoratimimus DNA and Brevibacterium epidermidis DNA were abundant in PL and Fusicatenibacter saccharivorans and P. argentinensis DNA were most often detected in BC (Fig 4).
Discussion
To our best knowledge, this is the first study aimed to characterize and compare the bacterial DNA populations of whole blood and its fractions between healthy calves and those with suspected sepsis.
Analysis of the microbiome revealed bacterial DNA in whole blood and its fraction in both groups; nevertheless, understanding its origin and role poses challenges. The finding of bacterial DNA in the blood samples from the S calves may indicate an alteration of the physiological barrier between body districts, with bacteria entering the bloodstream [23,26,27]. Conversely, the presence of bacterial DNA in the blood samples from the H calves is more difficult to explain. Previous studies detected bacterial traces from blood culture or gene sequencing in healthy patients, but whether this should be considered a proper microbiome, sporadic bacteremia, cell-free DNA or environmental contamination remains debated [10,28–30]. Therefore, further studies are needed to fully understand the meaning of this finding.
Not all the samples were suitable for sequencing; indeed, most RBC fractions from the S group could not be analyzed. One possible explanation is that alterations in erythrocyte metabolism and shape during sepsis [31] may have reduced blood sample quality or perhaps this type of sample requires specific precautions for preservation and transportation before undergoing analysis.
Intra-group microbiome analysis showed differences for both the S and the H group. Differences between blood fractions have also been reported in human medicine [9]; to our knowledge, however, similar studies have not been conducted on cattle. Such differences may arise from the different mechanisms by which bacterial DNA enters the bloodstream. For example, bacterial DNA fragments can be detected in the plasma in health and diseased patients due to apoptosis and necrosis even when blood cultures are negative [32]. Further, phagocytosis by white blood cells can be responsible for detectable bacterial DNA within the BC fraction [9]. Also, it has been demonstrated that red blood cells can capture and transport bacterial DNA, which explains the bacterial DNA detected in the RBC fraction in our samples [33]. What is not completely clear, however, is whether such differences result from a distribution of bacterial DNA from other factors such as blood separation technique [14,34]. When we compared the same blood fractions between the two groups, the α diversity indices indicated no significant differences, whereas Bray-Curtis dissimilarity disclosed differences between both groups for the PL and the WB fraction. Differential abundance analysis also showed significant differences. The relative abundance of the DNA of several bacterial species was statistically higher in the S than in the H group (Figs 3 and 4). It is possible that due to infection and inflammation bacteria entered the bloodstream from other body sites [35,36]. The higher abundance of P. aeruginosa DNA in the WB fraction from the S group could imply its role in sepsis, as it’s a well-known opportunistic bacterium [37].
In this fraction, there was also a high relative abundance of Cutibacterium acnes, Phascolabacterium sucinatutens, Roseburia intestinalis, and Staphylococcus. pasteuri DNA. Cutibacterium acnes, previously known as Propionibcterium acnes, a common inhabitant of the human skin and associated with pathological conditions [38], has been detected in the blood of septic human patients [13,39]. Moreover, since other studies found it in the rumen, placenta, and cow brain [40–42], we cannot exclude its involvement in the sepsis in the S calves. S. pasteuri has been detected in udders with mastitis [43], while the role of Roseburia intestinalis and Phascolabacterium succinatutens is unclear. The former appears to have a limiting effect on bacterial translocation owing to its butyrate production. Previous studies have reported it in cow gut, milk, and vaginal microbiomes [44–46]. The latter is an inhabitant of the ruminant digestive system and uses succinate for growth [47]. Therefore, their presence in the bloodstream could indicate bacterial translocation from the gut.
The E. coli DNA in the PL fraction was significantly more abundant in the S group. E. coli is a well-known microorganism in calf septicemia and our findings indicate that PL may be more efficient than the other blood fractions for detecting the DNA of this microorganism [48]. The relative abundance in PL of S. coagulans DNA was also high. This bacterium inhabits the mucous membranes and skin and has been associated with mastitis in some cases [49].
Though not commonly known to be pathogenic, the bacterial DNA of S. ilei and Fusobacterium mortiferum was significantly more often found in the BC fraction of the S group. S. ilei is a recently discovered microorganism [50] in humans. Further studies are needed to understand its role in cattle. Fusobacterium mortiferum has been isolated in cows with digital dermatitis and in the rumen [51,52]. Its presence in our sample might be attributed to bacterial translocation.
Several microorganisms had a higher relative abundance in the RBC fraction in the S group. Aside from the detection of the DNA of the potentially pathogenic bacterium Proteus mirabilis, the other bacterial DNA detected in RBC are opportunistic microorganisms are rarely studied in cattle [53,54]. However, because RBC could be analyzed for only three animals, a larger sample is needed to understand the role of these bacteria in this fraction.
The sequencing data indicate that the abundance of bacterial DNA from different species can differ in whole blood and its fractions. In our study, pathogenic bacteria (e.g., E. coli) was more often found in the PL than in the other fractions or whole blood. Nevertheless, analysis of PL alone would results in a loss of data on bacterial DNA in the bloodstream. While the underlying reasons for this variability remain unclear, our findings are shared by previous studies on humans which did not clarify which blood fraction can give a comprehensive overview of bacterial DNA in the bloodstream. [13,32,55].
E. coli was detected by blood culture in only two calves, while its DNA was significantly more abundant in the PL of 8 S calves compared to the H group. The origin of bacterial DNA in PL warrants further study, therefore our findings are insufficient to indicate whether viable E. coli was present even when the blood culture was negative. Studies in human medicine suggest, however, that the microbial DNA levels detected at sequencing may reflect the dynamics of infection [32,56], even when the microorganism is not detected by culture, and therefore could indicate a role for E. coli in clinical status. Further studies are needed to understand whether analysis of PL could help to evaluate the infection dynamics.
Several uncommon pathogens were isolated at blood culture in two calves. S. uberis is a relatively uncommon environmental pathogen affecting calves. Its involvement in sepsis should not be underestimated since it’s related to udder diseases and has also been detected in the blood of adult cows [36,57]. Similarly, P. stuartii is an environmental microorganism that inhabits cattle intestines; its detection in blood is unusual [58]. In human medicine, it has been detected in septic patients [59]. In our sample, the relative abundance did not significant differ between the S and H groups; therefore, we believe these bacteria might have been contaminants.
Despite we applied the commonly described techniques to perform sterile blood draws, it must be considered that there is a lack of literature describing the optimal procedures for performing a sterile blood sampling, in bovine, on field condition [8]. Pas et al. [8] reported a 14.90% contamination rate in blood cultures from calves although the sampling site had been prepped. While skin disinfection reduces the microbial population originating from cow hair and skin [60,61], routine procedures may not be sufficient to prevent contamination of blood cultures. In humans, repeated surgical site preparation before surgery was found to reduce skin bacterial loads [62]. Given the difficulty of maintaining a proper asepsis during on-field sampling, further studies should evaluate the effectiveness of a second surgical scrub, performed between the two sampling, on limiting the contamination of blood culture. Other factors were likely also at play in the differences between the blood culture results and the 16S rRNA gene sequencing analysis. Indeed, both clinical evaluation and blood culture have their limitations. Currently available clinical evaluation scores are imperfect tools for detecting sepsis in ruminants. The SIRS score can be influenced by the behavioral response of the patient [2,63]. In human medicine, the sepsis-related organ failure assessment (SOFA) score [4] appears efficient in the clinical evaluation of sepsis; however, it entails assessment of multiple parameters, such as fraction of inspired oxygen or arterial pressure, which are difficult to measure in buiatric practice. Moreover, this score has not yet been validated in ruminants. In the present study we sought to overcome the limitations of the SIRS score by including only critically ill calves [1]. The bacteremia score has a sensitivity and a specificity of 75% and 71%, respectively [22], which may explain why some calves tested negative at blood culture despite having a high score.
Blood culture has its limitations, too [5,6]. A low bacterial load can produce a negative culture. 16S rRNA sequencing has drawbacks as well: it does not discern between live and dead microorganisms nor it allow for antimicrobial susceptibility testing, which could be crucial for determining therapy effectiveness and assessing the antimicrobial resistance status of a herd.
None of these techniques is flawless and sepsis is not always associated with bacteremia. Indeed, the presence of cell-free bacterial DNA, detectable at 16S gene sequencing but not with culture-based methods, can trigger immune response in some cases [64]. If we wish to gain a deeper understanding of the mechanisms underlying sepsis, we will need to overcome the limitations of current diagnostic techniques. Lastly, not all the calves had the same site of infection. Owing to the small sample size, we were unable to analyze the effect of infection on the blood microbiome. Nevertheless, sepsis caused a shift in the bacterial DNA towards microorganisms with potential pathogenicity in the S calf group. Further, our findings are shared by studies in human medicine that suggest that the microbial population tends to shift during sepsis, regardless of whether there is one or multiple infection sites [12,65,66].
In conclusion, 16S rRNA gene sequencing holds promise for the evaluation of bacterial DNA in the blood fractions of S calves. The difference in the microbial population between PL and the other blood fractions suggests that PL may be a useful matrix for monitoring infection dynamics during sepsis. Further studies with larger samples are needed to better understand the role of bacteria in the bloodstream.
References
- 1. Pas ML, Bokma J, Lowie T, Boyen F, Pardon B. Sepsis and survival in critically ill calves: Risk factors and antimicrobial use. J Vet Intern Med. 2023;37(1):374–89. pmid:36562487
- 2. Bonelli F, Meucci V, Divers TJ, Boccardo A, Pravettoni D, Meylan M, et al. Plasma procalcitonin concentration in healthy calves and those with septic systemic inflammatory response syndrome. Vet J. 2018;234:61–5. pmid:29680395
- 3. House JK, Smith GW, Gunn AA, McGuirk SM, Izzo M. Manifestations and Management of Disease in Neonatal Ruminants. Large Animal Internal Medicine. Elsevier; 2019.
- 4. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315(8):801–10. pmid:26903338
- 5. Fecteau G, Smith BP, George LW. Septicemia and meningitis in the newborn calf. Vet Clin North Am Food Anim Pract. 2009;25(1):195–208, vii–viii. pmid:19174289
- 6. Hasman H, Saputra D, Sicheritz-Ponten T, Lund O, Svendsen CA, Frimodt-Møller N, et al. Rapid whole-genome sequencing for detection and characterization of microorganisms directly from clinical samples. J Clin Microbiol. 2014;52(1):139–46. pmid:24172157
- 7. Giancola S, Hart KA. Equine blood cultures: Can we do better?. Equine Veterinary Journal. 2022;55(4):584–92.
- 8. Pas ML, Boyen F, Castelain D, Chantillon L, Paepe D, Pille F, et al. Bayesian evaluation of sensitivity and specificity of blood culture media and hypoglycemia in sepsis-suspected calves. J Vet Intern Med. 2024;38(3):1906–16. pmid:38526076
- 9. Païssé S, Valle C, Servant F, Courtney M, Burcelin R, Amar J, et al. Comprehensive description of blood microbiome from healthy donors assessed by 16S targeted metagenomic sequencing. Transfusion. 2016;56(5):1138–47. pmid:26865079
- 10. Scarsella E, Zecconi A, Cintio M, Stefanon B. Characterization of Microbiome on Feces, Blood and Milk in Dairy Cows with Different Milk Leucocyte Pattern. Animals (Basel). 2021;11(5):1463. pmid:34069719
- 11. Mandal RK, Jiang T, Al-Rubaye AA, Rhoads DD, Wideman RF, Zhao J, et al. An investigation into blood microbiota and its potential association with Bacterial Chondronecrosis with Osteomyelitis (BCO) in Broilers. Sci Rep. 2016;6:25882. pmid:27174843
- 12. Szabó BG, Kiss R, Makra N, Pénzes K, Vad E, Kamotsay K, et al. Composition and changes of blood microbiota in adult patients with community-acquired sepsis: A pilot study from bench to bedside. Front Cell Infect Microbiol. 2022;12:1067476. pmid:36583109
- 13. Watanabe N, Kryukov K, Nakagawa S, Takeuchi JS, Takeshita M, Kirimura Y, et al. Detection of pathogenic bacteria in the blood from sepsis patients using 16S rRNA gene amplicon sequencing analysis. PLoS One. 2018;13(8):e0202049. pmid:30110400
- 14. James WA, Ogunrinde E, Wan Z, Kamen DL, Oates J, Gilkeson GS, et al. A Distinct Plasma Microbiome But Not Gut Microbiome in Patients With Systemic Lupus Erythematosus Compared to Healthy Individuals. J Rheumatol. 2022;49(6):592–7. pmid:35169056
- 15. Szabó BG, Kiss R, Makra N, Pénzes K, Vad E, Kamotsay K, et al. Composition and changes of blood microbiota in adult patients with community-acquired sepsis: A pilot study from bench to bedside. Front Cell Infect Microbiol. 2022;12:1067476. pmid:36583109
- 16. Lefterova MI, Suarez CJ, Banaei N, Pinsky BA. Next-Generation Sequencing for Infectious Disease Diagnosis and Management: A Report of the Association for Molecular Pathology. J Mol Diagn. 2015;17(6):623–34. pmid:26433313
- 17. Ferrini S, Grego E, Ala U, Cagnotti G, Valentini F, Di Muro G, et al. Feasibility of 16S rRNA sequencing for cerebrospinal fluid microbiome analysis in cattle with neurological disorders: a pilot study. Vet Res Commun. 2023;47(2):373–83. pmid:35759164
- 18. Nicola I, Cerutti F, Grego E, Bertone I, Gianella P, D’Angelo A, et al. Characterization of the upper and lower respiratory tract microbiota in Piedmontese calves. Microbiome. 2017;5(1):152. pmid:29157308
- 19. Mee JF. Invited review: Bovine neonatal morbidity and mortality-Causes, risk factors, incidences, sequelae and prevention. Reprod Domest Anim. 2023;58 Suppl 2:15–22. pmid:37128970
- 20.
The European Agency for the Evaluation of Medicinal Product. Guideline on Good Clinical Practices. Emea. 2000. p. 0–29.
- 21. Trefz FM, Feist M, Lorenz I. Hypoglycaemia in hospitalised neonatal calves: Prevalence, associated conditions and impact on prognosis. Vet J. 2016;217:103–8. pmid:27810199
- 22. Fecteau G, Paré J, Van Metre DC, Smith BP, Holmberg CA, Guterbock W, et al. Use of a clinical sepsis score for predicting bacteremia in neonatal dairy calves on a calf rearing farm. Can Vet J. 1997;38(2):101–4. pmid:9028593
- 23. Garcia J, Pempek J, Hengy M, Hinds A, Diaz-Campos D, Habing G. Prevalence and predictors of bacteremia in dairy calves with diarrhea. J Dairy Sci. 2022;105(1):807–17. pmid:34656356
- 24. Bolyen E, Rideout JR, Dillon MR, Bokulich NA, Abnet CC, Al-Ghalith GA, et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat Biotechnol. 2019;37(8):852–7. pmid:31341288
- 25. Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: High-resolution sample inference from Illumina amplicon data. Nat Methods. 2016;13(7):581–3. pmid:27214047
- 26. Amachawadi RG, Nagaraja TG. Pathogenesis of Liver Abscesses in Cattle. Vet Clin North Am Food Anim Pract. 2022;38(3):335–46. pmid:36243456
- 27. Scarsella E, Meineri G, Sandri M, Ganz HH, Stefanon B. Characterization of the Blood Microbiome and Comparison with the Fecal Microbiome in Healthy Dogs and Dogs with Gastrointestinal Disease. Vet Sci. 2023;10(4):277. pmid:37104432
- 28. Mtshali K, Khumalo ZTH, Kwenda S, Arshad I, Thekisoe OMM. Exploration and comparison of bacterial communities present in bovine faeces, milk and blood using 16S rRNA metagenomic sequencing. PLoS One. 2022;17(8):e0273799. pmid:36044481
- 29. Gosiewski T, Szała L, Pietrzyk A, Brzychczy-Włoch M, Heczko PB, Bulanda M. Comparison of methods for isolation of bacterial and fungal DNA from human blood. Curr Microbiol. 2014;68(2):149–55. pmid:24026449
- 30. Tan CCS, Ko KKK, Chen H, Liu J, Loh M, SG10K_Health Consortium, et al. No evidence for a common blood microbiome based on a population study of 9,770 healthy humans. Nat Microbiol. 2023;8(5):973–85. pmid:36997797
- 31. Bateman RM, Sharpe MD, Singer M, Ellis CG. The Effect of Sepsis on the Erythrocyte. Int J Mol Sci. 2017;18(9):1932. pmid:28885563
- 32. Grumaz S, Stevens P, Grumaz C, Decker SO, Weigand MA, Hofer S, et al. Next-generation sequencing diagnostics of bacteremia in septic patients. Genome Med. 2016;8(1):73. pmid:27368373
- 33. Damgaard C, Magnussen K, Enevold C, Nilsson M, Tolker-Nielsen T, Holmstrup P, et al. Viable bacteria associated with red blood cells and plasma in freshly drawn blood donations. PLoS One. 2015;10(3):e0120826. pmid:25751254
- 34. Chen D, He J, Li J, Zou Q, Si J, Guo Y, et al. Microbiome and Metabolome Analyses Reveal Novel Interplay Between the Skin Microbiota and Plasma Metabolites in Psoriasis. Front Microbiol. 2021;12:643449. pmid:33796091
- 35. Jeon SJ, Cunha F, Vieira-Neto A, Bicalho RC, Lima S, Bicalho ML, et al. Blood as a route of transmission of uterine pathogens from the gut to the uterus in cows. Microbiome. 2017;5(1):109. pmid:28841911
- 36. Brennecke J, Falkenberg U, Wente N, Krömker V. Are Severe Mastitis Cases in Dairy Cows Associated with Bacteremia?. Animals (Basel). 2021;11(2):410. pmid:33562677
- 37. Lotfollahzadeh S, Salehi TZ, Esmatabadi MZ, Ramezankhani M, Tamai IA, Shokri A, et al. Microbiological study of the auditory canal in dairy calves with otitis media. Microb Pathog. 2024;188:106547. pmid:38296119
- 38. Scholz CFP, Kilian M. The natural history of cutaneous propionibacteria, and reclassification of selected species within the genus Propionibacterium to the proposed novel genera Acidipropionibacterium gen. nov., Cutibacterium gen. nov. and Pseudopropionibacterium gen. nov. Int J Syst Evol Microbiol. 2016;66(11):4422–32. pmid:27488827
- 39. Gull T. Bacterial Causes of Intestinal Disease in Dairy Calves: Acceptable Control Measures. Vet Clin North Am Food Anim Pract. 2022;38(1):107–19. pmid:35219479
- 40. Boman J, Nilson B, Sunnerhagen T, Rasmussen M. True infection or contamination in patients with positive Cutibacterium blood cultures-a retrospective cohort study. Eur J Clin Microbiol Infect Dis. 2022;41(7):1029–37. pmid:35612767
- 41. Park Y-J, Kang G-U, Jeong M, Singh V, Kim J, Lee K, et al. Bacterial Profiles of Brain in Downer Cattle with Unknown Etiology. Microorganisms. 2022;11(1):98. pmid:36677388
- 42. Lyons JA, Bemis DA, Bryant MJ, Kania SA, Abd-Eldaim M, Newman SJ. Isolation of Propionibacterium acnes from a case of placentitis and abortion in a cow. J Vet Diagn Invest. 2009;21(2):274–7. pmid:19286514
- 43. Koniarová I. Biochemical and physiologic properties of strains of Propionibacterium acnes isolated from the rumen of calves and lambs. Vet Med (Praha). 1993;38(1):43–52. Available: https://europepmc.org/article/med/8475617 pmid:8475617
- 44. Condas LAZ, De Buck J, Nobrega DB, Carson DA, Naushad S, De Vliegher S, et al. Prevalence of non-aureus staphylococci species causing intramammary infections in Canadian dairy herds. J Dairy Sci. 2017;100(7):5592–612. pmid:28527793
- 45. Fan P, Nelson CD, Driver JD, Elzo MA, Peñagaricano F, Jeong KC. Host genetics exerts lifelong effects upon hindgut microbiota and its association with bovine growth and immunity. ISME J. 2021;15(8):2306–21. pmid:33649551
- 46. Tardón DC, Hoffmann C, Santos FCR, Decaris N, Pinheiro FA, Queiroz LL, et al. Relationships among Indicators of Metabolism, Mammary Health and the Microbiomes of Periparturient Holstein Cows. Animals (Basel). 2021;12(1):3. pmid:35011109
- 47. Zhao C, Bao L, Qiu M, Wu K, Zhao Y, Feng L, et al. Commensal cow Roseburia reduces gut-dysbiosis-induced mastitis through inhibiting bacterial translocation by producing butyrate in mice. Cell Rep. 2022;41(8):111681. pmid:36417859
- 48. Conteville LC, da Silva JV, Andrade BGN, Cardoso TF, Bruscadin JJ, de Oliveira PSN, et al. Rumen and fecal microbiomes are related to diet and production traits in Bos indicus beef cattle. Front Microbiol. 2023;14:1282851. pmid:38163076
- 49.
Gyles CL, Fairbrother JM. Escherichia Coli. Pathogenesis of Bacterial Infections in Animals. Wiley. 2010. 267–308. https://doi.org/10.1002/9780470958209.ch15
- 50. de Freitas Guimarães F, Nóbrega DB, Richini-Pereira VB, Marson PM, de Figueiredo Pantoja JC, Langoni H. Enterotoxin genes in coagulase-negative and coagulase-positive staphylococci isolated from bovine milk. J Dairy Sci. 2013;96(5):2866–72. pmid:23477822
- 51. Hyun D-W, Lee J-Y, Kim M-S, Shin N-R, Whon TW, Kim KH, et al. Pathogenomics of Streptococcus ilei sp. nov., a newly identified pathogen ubiquitous in human microbiome. J Microbiol. 2021;59(8):792–806. pmid:34302622
- 52. Caddey B, Orsel K, Naushad S, Derakhshani H, De Buck J. Identification and Quantification of Bovine Digital Dermatitis-Associated Microbiota across Lesion Stages in Feedlot Beef Cattle. mSystems. 2021:6.
- 53. Schwarz C, Mathieu J, Gomez JL, Miller MR, Tikhonova M, Nagaraja TG, et al. Unexpected finding of Fusobacterium varium as the dominant Fusobacterium species in cattle rumen: potential implications for liver abscess etiology and interventions. J Anim Sci. 2023;101:skad130. pmid:37104065
- 54. McMahon DJ, Bowen IB, Green I, Domek M, Oberg CJ. Growth and survival characteristics of Paucilactobacillus wasatchensis WDCO4. J Dairy Sci. 2020;103(10):8771–81. pmid:32747094
- 55. Liu Q, Liu H-C, Yang L-L, Xin Y-H. Flavobacterium franklandianum sp. nov. and Flavobacterium gawalongense sp. nov., isolated from glaciers on the Tibetan Plateau. Int J Syst Evol Microbiol. 2021;71(7):10.1099/ijsem.0.004868. pmid:34228609
- 56. Wu C, Yu X, Gai W, Liu Y, Qi Y, Zheng Y, et al. Diagnostic value of plasma and blood cells metagenomic next-generation sequencing in patients with sepsis. Biochem Biophys Res Commun. 2023;683:149079. pmid:37871447
- 57. Fida M, Wolf MJ, Hamdi A, Vijayvargiya P, Esquer Garrigos Z, Khalil S, et al. Detection of Pathogenic Bacteria From Septic Patients Using 16S Ribosomal RNA Gene-Targeted Metagenomic Sequencing. Clin Infect Dis. 2021;73(7):1165–72. pmid:33893492
- 58. Krebs I, Zhang Y, Wente N, Leimbach S, Krömker V. Bacteremia in Severe Mastitis of Dairy Cows. Microorganisms. 2023;11(7):1639. pmid:37512812
- 59. Waldhalm DG, Meinershagen WA, Frank FW. Providencia stuartii as an etiologic agent in neonatal diarrhea in calves. Am J Vet Res. 1969;30(9):1573–5. pmid:5817009
- 60. George EA, Kornik R, Robinson-Bostom L. Providencia stuartii septic vasculitis. JAAD Case Rep. 2020;6(5):422–5. pmid:32382635
- 61. Marchionatti E, Constant C, Steiner A. Preoperative skin asepsis in bovine surgery: an outcome-blinded 3-arm randomized clinical trial under non-sterile operating room conditions. Front Vet Sci. 2024;11:1446649. pmid:39711800
- 62. Blonna D, Allizond V, Bellato E, Banche G, Cuffini AM, Castoldi F, et al. Single versus Double Skin Preparation for Infection Prevention in Proximal Humeral Fracture Surgery. Biomed Res Int. 2018:8509527.
- 63. Yanar KE, Eren E, Aktaş MS, Eroğlu MS, Kandemir Ö, Aydın G. Prognostic potential of inflammatory markers, oxidative status, thrombocyte indices, and renal biochemical markers in neonatal calf diarrhoea-induced systemic inflammatory response syndrome. Vet Immunol Immunopathol. 2023;265:110680. pmid:37980800
- 64. Kaewduangduen W, Visitchanakun P, Saisorn W, Phawadee A, Manonitnantawat C, Chutimaskul C, et al. Blood Bacteria-Free DNA in Septic Mice Enhances LPS-Induced Inflammation in Mice through Macrophage Response. Int J Mol Sci. 2022;23(3):1907. pmid:35163830
- 65. Gosiewski T, Ludwig-Galezowska AH, Huminska K, Sroka-Oleksiak A, Radkowski P, Salamon D, et al. Comprehensive detection and identification of bacterial DNA in the blood of patients with sepsis and healthy volunteers using next-generation sequencing method - the observation of DNAemia. Eur J Clin Microbiol Infect Dis. 2017;36(2):329–36. pmid:27771780
- 66. Wang C, Li Q, Tang C, Zhao X, He Q, Tang X, et al. Characterization of the blood and neutrophil-specific microbiomes and exploration of potential bacterial biomarkers for sepsis in surgical patients. Immun Inflamm Dis. 2021;9(4):1343–57. pmid:34288545