Optimization of bacterial separation techniques in sepsis for third-generation sequencing

doi:10.21203/rs.3.rs-1514449/v1

Download PDF

Article

Optimization of bacterial separation techniques in sepsis for third-generation sequencing

https://doi.org/10.21203/rs.3.rs-1514449/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background:

Sepsis is a common serious condition worldwide and diagnostics are still slow. Nanopore sequencing is a promising tool in point of care medicine. However, high ratio of host to bacterial DNA decreases utility of nanopore. Fast, cheap, and easy techniques to separate host from bacterial DNA are required.

Aim:

The aim was to compare three different separation methods of blood for bacterial DNA enrichment and isolation with subsequent products readily suitable for nanopore sequencing.

Methods:

Blood was collected from healthy subjects and spiked with either gram-negative or gram-positive bacteria. Three different separation methods alongside negative control were performed. The performance of each method was evaluated by RT-qPCR of 16S gene and B2 microglobulin.

Results:

16s RT-qPCR showed similar Ct value for tested methods in comparison to Standard method.

However, the host DNA depletion was significantly higher for the centrifugal method compared to others. Reproducibility, based on variability measured by the standard deviation and RT-qPCR results, was highest using the centrifugation method compared to the other three methods.

Conclusion:

From all tested methods, centrifugation outperformed other basic isolation techniques considering turnaround time, reliability, and effectiveness. Further investigation is needed to confirm our promising results in a real-life scenario.

Patients with sepsis represent a substantial number of hospital admissions, including ICU admissions. Sepsis and its more severe form, septic shock, has a 25–50% mortality rate [1]. Additionally, sepsis poses a huge economic burden contributing to 13% of total hospital costs with only 3.6% of total hospital admissions [2]. Although research has unraveled more pathomechanisms lately, there is still no specific treatment. The mainstay is appropriate antibiotic therapy and supportive therapy (eg. intravenous therapy, vasoactive therapy).

The current gold standard for choosing the right antibiotic therapy is based on phenotypic assay; however, this process has an immense turnaround time of 24–48 hours. While awaiting results, the clinicians have to start empiric antibiotic therapy. Approximately 20% of the time, this therapy is inappropriate, and it does not cover causative pathogens [3]. Besides treatment failure, broad-spectrum antibiotic therapy increases the number of multiresistant microorganisms.

With recent advances in molecular biology, faster methods would be expected to be currently in use. While several PCR methods are readily available, these cover only some specific pathogens and some of the antibiotic resistance genes [4]. Sanger sequencing and next-generation sequencing technologies both provide sufficient data output and accuracy but require skilled laboratory personnel and a bioinformatician for data analysis. Further, the slow throughput with high initial and running costs are making the analysis/system ineffective. Recent progress in genetic sequencing technology such as nano-pore sequencing, provides a tempting approach for faster sepsis diagnostics that includes all potential pathogens and its antibiotic resistances. Size of the device, time to the result, and overall costs are substantially lower compared to the previous generations, thus making the nano-pore sequencing more promising for sepsis diagnostics. The limiting step in this workflow is the amount of pathogen DNA. The exact amount of bacterial cells in the bloodstream in sepsis is not known, but 10 colony forming units (CFU) per ml of blood poses a risk for sepsis development [5]. This calls for the efficient separation of bacterial cells from host cells and represents a major limitation why third-generation sequencing is not a widespread method of pathogen identification.

Currently, there are several methods investigated for microbial separation from blood including mechanical filtration, magnetic cell sorting and/or microfluidic methods [6, 7]. These, methods require advanced laboratory equipment and skilled personnel which increase the economic burden and so far, are not suitable for potential point of care testing. In our research, we focused on the adaptation and development of relatively easy and rapid separation methods - centrifugal, chemical and enzymatic. The principle of the centrifugal method is sequential centrifugation of septic blood collected into a serum collection tube. First centrifugation removes the eukaryotic cells under the separation gel [8]. The second centrifugation step pellets the bacteria. based on different sedimentations speeds between microbial and eukaryotic cells to obtain a nearly pure fraction of bacterial cells/pathogens. Polaris is the chemical method invented by Dobbelaer I. et al., 2012 [9]. In this case, eukaryotic cells and bacterial cells differ in the composition of cell membranes and with the presence of cell walls in bacteria. Selective lysis of eukaryotic cells’ membrane is performed by the alkaline ionic surfactant solution that interacts with the membrane. Then, DNA is released to the solution and is degraded by an alkaline environment whereas the majority of the bacteria remain intact and easy to collect. The enzymatic method called MolYsis Complete5 uses a similar approach as Polaris in lysis of eukaryotic cell walls but instead of ionic surfactant, the chaotropic buffer is used. Subsequently, released DNA is degraded by a DNase [10].

Therefore, the aim of our research was to optimize and to compare the effectiveness of the described methods with respect to standard DNA isolation to be used potentially in the future as part of the pipeline in sepsis diagnostics by nanopore sequencing.

Agar plates to verify the correct spiking concentration of blood.

Agar plate was inoculated with 10 000 to 10 CFU /ml (Figure 1; from upper line to bottom line). In the first line, the concentration was estimated to be approximately 10 000 CFU /ml (Fig.1-A1), the second line approximately 1000 CFU /ml (Fig.1-A2), the third line 100 CFU /ml (Fig.1-A3), and the fourth line 20 CFU /ml (Fig.1-A4). Another agar was inoculated with blood samples contaminated with bacteria to verify successful spiking (Fig. 1-B). Colonies were formed on agar plates inoculated with spiked blood from serum collection tube and K₃EDTA collection tube (Fig.1-B1, B4). Negative control remained sterile and no bacterial growth was observed (Fig.1-B2, B3). An overview is shown in Figure 1.

DNA quantity and quality analysis

The concentration of DNA was significantly lower in Centrifugation and Polaris separation methods than in Standard method without any separation (Figure 2; p<0.001). To a lesser extent, the isolated concentration of DNA in Molzym was significantly lower (p<0.01) than the Standard method. The average concentration of DNA with only Standard method was 57.09 ± 25.07 ng/ml compared to 3.38 ± 1.54 ng/ml with Centrifugation, 7.87 ± 10.62 ng/ml with Polaris, and 15.55 ± 12.62 ng/ml with Molzym (Figure 2). Standard method performed significantly better in the 260/230 ratio (Figure 3) than the other three methods. Among separation methods, there were no significant differences in the 260/230 and 260/280 ratios, respectively. However, the Centrifugation and Polaris method showed relatively high variability (Figure 4).

RT-qPCR analysis of 16s E. coli DNA of individual isolation techniques

To evaluate efficacy of individual isolation techniques, RT-qPCR for 16S DNA detection was performed. Both Centrifugation and Molzym methods showed comparable Ct values that were significantly lower than Standard method. Moreover, the Centrifugation method showed the least amount of variability (Figure 5). The mean Ct value of Centrifugation, Molzym, Polaris, and Standard methods were 29.37 ± 0.39, 29.38 ± 0.92, 29.98 ± 0.67, 30.64 ± 0.95, respectively. The Ct value difference between the spiked sample and its non-spiked control showed that Molzym performed significantly better (p<0.05) when compared to the Qiagen and Polaris methods. Centrifugation method and Molzym performed similarly with no observed difference in Ct values (Figure 6).

RT-qPCR analysis of 16s S.Aureus DNA of individual isolation techniques

Average Ct value for 16s S.aureus was significantly different between Molzym and all three other methods. (Figure 7). The average CT value for negative template control (NTC) was 32.47. The mean CT value of Centrifugation, Molzym, Polaris, and Standard method were 31.25 ± 0.77, 29.24 ± 0.78, 30.98 ± 0.78, 31.45 ± 0.63, respectively. The difference between a spiked sample and its non-spiked control showed a significantly higher difference in Molzym methods than in the other three methods (Figure 8).

RT-qPCR analysis of 16s S.Aureus DNA of individual isolation techniques with lysozyme addition

There were no significant differences in average Ct value within the four methods (Figure 9) after addition of lysozyme. The mean Ct value for NTC was 36.06. The mean Ct values for Centrifugation, Molzym, Polaris, and Standard methods were 30.93 ± 2.52, 31.47 ± 1.34, 30.52 ± 1.82, and 31.73 ± 2.03, respectively. Differences between the spiked sample and its non-spiked control (Figure 10) showed no significant differences among methods; however, a trend between Centrifugation and the Molzym method can be seen (p=0.075).

RT-qPCR analysis of B2 microglobulin DNA of individual isolation techniques

Significantly different average Ct values were observed between Centrifugation and Standard method, Centrifugation, and Molzym method, Polaris and Standard method (Figure 11). The mean Ct value for NTC was 37.47. The mean Ct values for Centrifugation, Molzym, Polaris, and Standard methods were 26.26 ± 1.31, 21.89 ± 2.2, 24.44 ± 3.33, and 20.48 ± 0.59, respectively.

Sepsis is a life-threatening condition with a relatively high prevalence. The fast and precise treatment is vital for mortality and morbidity reduction. However, current diagnostic tools are inadequate. This opens a window for newer approaches, namely 3rd generation sequencing. In order to utilize, efficient separation of bacterial DNA from host DNA in blood is the first essential step. Several commercial methods and protocols are currently available, and these were compared in effectiveness to separate bacterial DNA from host DNA in spiked blood samples.

Two agar plates were inoculated after spiking the blood cultures to obtain 100 CFU /ml concentration. The first agar plate was inoculated with 10⁴– 10¹ dilutions of prepared cultures and the second with the spiked blood samples. This served as retrospective verification if the samples were correctly spiked with the desired 100 CFU /ml concentration of bacteria. The CFUs formed on the first agar plate proved the spiking of the samples with 100 CFU /ml. The CFUs formed on the second agar plate proved that spiked blood samples in the serum collection tubes and the K₃EDTA collection tubes were spiked with Escherichia coli / Staphylococcus aureus. More CFUs were formed in the spiked blood samples in the serum collection tubes compared to K₃EDTA tubes. Spiked samples in K₃EDTA tubes had less CFU on the agar plate, which could be caused by the possible inhibiting effect of K₃EDTA on the growth of the bacteria [11]. On the contrary, K₃EDTA tubes had the highest recovery of bacteria compared to other anticoagulants such as citrate or heparin. Furthermore, blood obtained from K₃EDTA vacutainers contained a high number of platelets compared to the others where little or no platelets were observed [12]. However, the relationship between platelets and bacteria recovery needs to be further investigated.

DNA quantification showed that the Standard method without any separation technique had the highest DNA concentration, followed by the Molzym method. In the case of the Standard method, this is most probably due to the high abundance of eukaryotic DNA in the sample. Molzym had significantly lower DNA concentration isolated compared to the Standard method; however, it was still significantly higher than the Centrifugation and Polaris method. In the Molzym method, DNase eliminates eukaryotic DNA, however the DNase activity may be inhibited or its activity is not sufficient for the vast amount of host DNA. The low DNA concentrations in Centrifugation and Polaris methods are due to the significant removal of eukaryotic DNA. Both methods early in the protocol adapt harsh mechanical or chemical treatment for such removal. This was confirmed by beta-2-microglobulin RT-qPCR. The Centrifugation method was consistently (coefficient of variation 4.98%) able to remove the majority of host DNA compared to the other two methods. Polaris also removed a substantial amount of host DNA, however with high variability (coefficient of variation 13.6%), which made this method less reliable. We also observed a negative correlation between the amount of DNA isolated with Ct value of beta-2-microglobulin RT-qPCR.

Indeed, apart from the elimination of eukaryotic DNA and the final DNA concentration, purity is one of the major factors that influence the nanosequnecing. In our study, we observed significantly higher purity in the Qiagen method than in every other method. However, these values are not decisive since several washing steps are involved in further downstream processes. Moreover, extremely low 260/230 ratios can be caused by low DNA concentration approaching 2 ng/ml at which every interference with other substances present in the eluate may alternate the purity. These concentrations were typical in Centrifugation and Polaris methods.

RT-qPCR for 16s gene for E. coli showed the highest sensitivity for bacteria detection (lowest Ct value) for Centrifugation and Molzym methods. Nonetheless, the Centrifugation method exhibited almost a two-fold lower coefficient of variation than Molzym method (1.34% vs. 3.13%). Apart from this variability, the Centrifugation and Molzym methods had similar efficiency compared to the other two methods.

RT-qPCR for 16s gene for S.aureus showed that Molzym would be the method of choice since the other three methods had similar Ct values compared with unspiked blood (negative control). We assume that the Qiagen kit did not sufficiently break the cell wall of S.aureus for DNA to be released, while Molzym did. Based on this observation, another experiment was conducted with addition of lysozyme during Qiagen isolation. After the addition of lysozyme, the efficacy of other methods increased and reached Molzym’s effectiveness. However, all methods seem to have suffered from high variability, which could have been caused by inconsistent bacterial cell wall lysis. Comparison of the difference between Ct value of the method and its negative control (non-spiked blood with the same treatment) showed decreased variability. This was true mainly for the Centrifugation method along with a trend of superiority compared to the Molzym method.

Turnaround time (TAT) of each method should also be considered as this is one of the factors in the successful treatment of sepsis. Centrifugation takes 20 minutes, Polaris 30 minutes, and Molzym takes around 80 minutes, making the centrifugation method the shortest. The Centrifugation method is also the least laborious. The other two methods need fresh preparation of chemicals (Polaris), is reliant on enyzme which may produce variable results due to instability (Molzym), or an error may occur during separation (Molyzm), leading to problems in downstream processes, ultimately leading to false-negative results in sepsis diagnosis.

We observed several tendencies, which should be evaluated further on a larger scale. Also, the concentration of the spiked bacteria in the blood samples could be too low for analysis to be successful; however, simulation of clinical scenarios is vital as sepsis could evolve from as low as 10 CFU /ml of blood. In a similar study led by Anne J. M. Loonen et al. (2013) [9], Molzym and Polaris separation methods were compared. They investigated the DNA separation at the concentrations from 1 CFU/ml to 1000 CFU/ml. The higher detection rate of the bacterial DNA was observed at 1000 CFU /ml. In addition, they studied the volume of the analyzed blood. The average Ct values were lower with increased blood sample volume so that the separation methods could separate more of the bacterial DNA. In comparison to our results, Molzym was a more sensitive method than Polaris but much less reliable. In another study led by H.-P. Horz et al. (2010) [13], the authors demonstrated that the Molzym method can eliminate the eukaryotic (host) DNA. Interestingly, the eukaryotic DNA was reduced by more than 90% in all 16 samples. This contrasts with our findings that Molzym was the least efficient method in depleting the host DNA. We consider these differences as a drawback of the Molzym method because of the long TAT with a more complicated workflow that creates a space for sample degradation and human error. However, a study led by Bernhardt et al. (1991) observed that the Centrifugation method could efficiently separate bacteria from blood, which agrees with the result of our study [14]. Similarly, Buchanan et al., performed mechanical separation of bacteria on a spinning disc, where they could rapidly separate as low as 200 to 6 CFU /ml, which agrees with our observation [12]. Another important consideration is the volume taken from the patient for bacteria identification. To detect as low as 10 CFU/ml, one needs 7 to 10 ml of whole blood volume for sufficient signal-to-noise antibiotic-resistant ratio to provide molecular identification [9].

There are several other techniques not tested in our condition. Magnetic beads capture, dielectrophoretic technology, or microfluidic devices are methods with high recovery of the bacteria but with high cost, trained personnel, and longer TAT. These were omitted as our aim was to create a simple, economically friendly, and on-site procedure for bacterial separation [9]. Considering TAT, reliability, and effectiveness, the Centrifugation method seems to surpass the other basic isolation techniques and should be considered in experiments where separation of eukaryotic from bacterial cells is needed for third-generation sequencing. Next, confirmation of our results on real samples of sepsis patients should be performed.

Bood acqusition and bacterial spiking procedure

The project was approved by the Ethical committee of University Hospital Bratislava, Ružinov, decision No. EK/206/2019. All involved participants provided informed consent. The research was conducted in accordance with the Declaration of Helsinki. The blood collection was performed by medical personnel from 4 healthy young volunteers. In total, 72 ml of blood from each healthy individual were collected into K₃EDTA blood collection tubes (Polaris method, Molzym method, Control sample) and serum collection tubes (Centrifugation method). The collected blood was aliquoted to obtain triplicates for each method. Non-spiked blood was used as a negative control. Escherichia coli CFT 073 and Staphylococcus aureus have been cultured overnight (approx. 20h) in LB broth (Sigma-Aldrich) prior to the spiking procedure. To determine the overall bacterial concentration, the optical density of bacterial suspension was measured at 600 nm (OD600). Afterward, the bacterial suspensions were diluted to concentrations from 10⁴ to 10¹ colony forming units per milliliter (CFU /ml). Subsequently, these diluted samples were plated on the LB agar, and incubated overnight for retrospective validation. One ml of 1000 CFU /ml of each bacterial suspension was further spiked into the 9 ml serum blood collection tubes, and 0.375 ml of each bacterial suspension was spiked into the 3 ml K₃EDTA blood collection tubes to achieve a final concentration of 100 CFU /ml of the whole sample. To simulate a real-life environment during which the sample needs to reach the laboratory, the samples were incubated for 30 minutes and administered for further processing (Figure 12). Spiked blood samples were plated on the LB agar and incubated overnight as a control for successful spiking process. All steps of the spiking process were performed under sterile conditions under the laminar flow hood.

Centrifugation

Spiked blood in the serum collection tube was centrifuged for 2 000 g for 10 minutes. Afterward, the supernatant was collected without disturbing the middle layer, placed into sterile 5 ml Eppendorf tubes, and centrifuged at 20 000 g for 10 minutes. Then, the supernatant was carefully discarded, and the pellet was resuspended in 200 μl of sterile PBS. Subsequently, DNA isolation was performed.

Molzym

For the Molzym DNA separation method, the whole treatment of the samples was performed according to the manufacturer’s instructions from the MolYsis ™ Complete5 kit (Molzym GmbH, Germany). Briefly, the host cells in a blood sample were lysed by adding a chaotropic buffer, while microbial cells remained unaffected. Then, the DNA released from host cells and dead cells were degraded by Molzym's MolDNase B. The suspension was further centrifuged. The pellet composed of microbial cells was then treated with BugLysis reagents to degrade cell walls of Gram-negative bacteria and Gram-positive bacteria and then digested by Proteinase K. The microbial DNA was extracted by using the spin column and then isolated by a specific buffer, 70% ethanol wash and final elution with deionized water. Finally, the isolation of DNA was carried out according to the manufacturer's instructions.

Polaris

The Polaris method was performed according to published study [9]. Briefly, spiked blood from the K₃EDTA collection tube was divided into 1 ml triplicates. Subsequently, 2 ml of Polaris lysis solution composed of 500 mM sodium bicarbonate and 1% Triton-X (Sigma-Aldrich, USA) were added. The mixture was vigorously mixed for 5 seconds and incubated at room temperature for 5 min. After incubation, a Polaris neutralizing solution composed of 1 M Tris-HCl was added, and tubes were centrifuged at 2800 g for 10 minutes. The supernatant was carefully discarded, and 1 ml of sterile PBS was added to the pellet, and centrifugation at 20 000 g for 10 minutes followed. After the last centrifugation step, 200 μl of sterile PBS was added to the pellet, and the suspension was ready for DNA extraction.

Standard treatment (Qiagen method)

Two hundred μl of the whole blood was treated with QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions.

Bacterial DNA isolation

Extraction of total DNA was carried out using Qiagen DNA Mini Kit (Qiagen, Hilden, Germany). Additionally, for the repeated experiment with S. aureus, a lysozyme enzyme solution (Sigma-Aldrich, USA) was added to the suspension prior to Qiagen isolation according to manfucaturer instrucions for DNA from bacteria isolation.

Bacterial and Genomic DNA quantification

Bacterial 16s RT-qPCR

TaqMan primers (V7-V9 regions of 16S rRNA) and the probe (Microsynth, Switzerland) were used to detect bacterial DNA (Supplementary Table S1). A twenty μl reaction mixture was composed of 1 μl of each primer (0.5 μM), 0.5 μl of the probe (0.25 μM), 10 μl of TaqMan Fast Advance Master Mix (AppliedBiosystems, USA), 3.5 μl of distilled water, and 4 μl of the template sample. RT-qPCR program was set up to pre-incubation at 50 °C for 2 minutes, initial denaturation at 90 °C for 10 minutes, followed by 40 cycles of denaturation at 95 °C for 15 seconds, annealing at 60 °C for 30 seconds and extension at 72 °C for 65 °C. Analysis was carried out on the Mastercycler Realplex 4 (Eppendorf, Germany).

Genomic RT-qPCR

For detection of eukaryotic DNA, Beta-2-microglobulin as a housekeeping gene was used (Supplementary Table S2). Ten μl reaction was set up, composed of 5 μl of SsoAdvanced universal SYBR® Green supermix (2x), 0.25 μl of forward and reverse primer (0.25 μM) (Bio-Rad, USA), 2 μl of PCR grade water, and 2.5 μl of template DNA. Cycling conditions were set up as follows: initial denaturation at 98 °C for 3 minutes followed by 40 cycles of denaturation at 98 °C for 15 seconds, annealing at 51 °C for 30 seconds, and extension at 60 °C for 30 seconds. After the end of the qPCR, all amplicons were melted to observe the melting curve of the present amplified products. The RT-qPCR analysis was performed on the Mastercycler Realplex 4 (Eppendorf, Germany).

Statistical analysis

All statistical analyses were performed using Graphpad prism 8.0.1 (La Jolla, California, USA). Non-parametric Kruskal-Wallis test with subsequent Dunn post-hoc correction tests was performed to analyze differences between the groups. P-value <0.05 was considered significant. Data are presented as mean +/- SD.

Author Contributions

RL,MT,TS performed the experiment and analyzed the data; RL,MT,BK,LT,JH designed the experiment; RL,MT,TS drafted the manuscript; BK,LT,JH co-drafted the manuscript; LT,JH secured funding; JH is the guarantor of this work and, as such, had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Acknowledgement

This study was supported by grants from the Slovak Research and Development Agency No. APVV-20-0472, APVV-18-0287 and Grant Agency of the Ministry of Education, Science, Research and Sport of the Slovak Republic No. VEGA 1/0573/21

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Additional Information

The authors declare no competing interests.

Vincent, J. L. et al. Assessment of the worldwide burden of critical illness: the intensive care over nations (ICON) audit. The Lancet. Respiratory medicine 2, 380–386, doi:10.1016/s2213-2600(14)70061-x (2014).
Paoli, C. J., Reynolds, M. A., Sinha, M., Gitlin, M. & Crouser, E. Epidemiology and Costs of Sepsis in the United States-An Analysis Based on Timing of Diagnosis and Severity Level. Critical care medicine 46, 1889–1897, doi:10.1097/ccm.0000000000003342 (2018).
Kumar, A. et al. Initiation of inappropriate antimicrobial therapy results in a fivefold reduction of survival in human septic shock. Chest 136, 1237–1248, doi:10.1378/chest.09-0087 (2009).
Kothari, A., Morgan, M. & Haake, D. A. Emerging technologies for rapid identification of bloodstream pathogens. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America 59, 272–278, doi:10.1093/cid/ciu292 (2014).
Yagupsky, P. & Nolte, F. S. Quantitative aspects of septicemia. Clinical microbiology reviews 3, 269–279, doi:10.1128/cmr.3.3.269 (1990).
Buchanan, C. M. et al. Rapid separation of very low concentrations of bacteria from blood. Journal of microbiological methods 139, 48–53, doi:10.1016/j.mimet.2017.05.004 (2017).
Han, J. Y., Wiederoder, M. & DeVoe, D. L. Isolation of intact bacteria from blood by selective cell lysis in a microfluidic porous silica monolith. Microsystems & Nanoengineering 5, 30, doi:10.1038/s41378-019-0063-4 (2019).
Matsui, A. et al. A Rapid ATP Bioluminescence-based Test for Detecting Levofloxacin Resistance Starting from Positive Blood Culture Bottles. Scientific Reports 9, 13565, doi:10.1038/s41598-019-49358-9 (2019).
Loonen, A. J. et al. Comparison of pathogen DNA isolation methods from large volumes of whole blood to improve molecular diagnosis of bloodstream infections. PLoS One 8, e72349, doi:10.1371/journal.pone.0072349 (2013).
Gebert, S., Siegel, D. & Wellinghausen, N. Rapid detection of pathogens in blood culture bottles by real-time PCR in conjunction with the pre-analytic tool MolYsis. The Journal of infection 57, 307–316, doi:10.1016/j.jinf.2008.07.013 (2008).
Cherian, L. M. et al. Effect of commercial nasal steroid preparation on bacterial growth. International forum of allergy & rhinology 9, 766–775, doi:10.1002/alr.22312 (2019).
Pitt, W. G. et al. Rapid separation of bacteria from blood-review and outlook. Biotechnology progress 32, 823–839, doi:10.1002/btpr.2299 (2016).
Horz, H. P., Scheer, S., Vianna, M. E. & Conrads, G. New methods for selective isolation of bacterial DNA from human clinical specimens. Anaerobe 16, 47–53, doi:10.1016/j.anaerobe.2009.04.009 (2010).
Bernhardt, M., Pennell, D. R., Almer, L. S. & Schell, R. F. Detection of bacteria in blood by centrifugation and filtration. Journal of clinical microbiology 29, 422–425, doi:10.1128/jcm.29.3.422-425.1991 (1991).

Supplementary Tables S1-S2 are not available with this version

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Optimization of bacterial separation techniques in sepsis for third-generation sequencing

Status:

Version 1

Abstract

Figures

Introduction

Results

Discussion

Materials And Methods

Declarations

Author Contributions

Acknowledgement

References

Supplemental Data

Additional Declarations

Status:

Version 1