Accepted Manuscript

Comparison of three methods for recovery of Brucella canis DNA

from canine blood samples

Maria Cryskely A. Batinga, Jaíne C. dos Santos, Julia T.R. Lima,

Maria Fernanda D. Bigotto, Kerstin Muner, Thalita Faita, Rodrigo
M. Soares, David A.V. da Silva, Trícia M.F.S. Oliveira, Helena L.
Ferreira, Jaqueline A. Diniz, Lara B. Keid

PII: S0167-7012(17)30229-4
DOI: doi: 10.1016/j.mimet.2017.08.019
Reference: MIMET 5232
To appear in: Journal of Microbiological Methods
Received date: 2 May 2017
Revised date: 28 August 2017
Accepted date: 29 August 2017

Please cite this article as: Maria Cryskely A. Batinga, Jaíne C. dos Santos, Julia T.R. Lima,
Maria Fernanda D. Bigotto, Kerstin Muner, Thalita Faita, Rodrigo M. Soares, David A.V.
da Silva, Trícia M.F.S. Oliveira, Helena L. Ferreira, Jaqueline A. Diniz, Lara B. Keid
, Comparison of three methods for recovery of Brucella canis DNA from canine blood
samples, Journal of Microbiological Methods (2017), doi: 10.1016/j.mimet.2017.08.019

This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.
 ACCEPTED MANUSCRIPT

Comparison of three methods for recovery of Brucella canis DNA from canine blood
samples

Maria Cryskely A. Batinga1, Jaíne C. dos Santos2, Julia T. R. Lima1, Maria Fernanda D.
Bigotto2, Kerstin Muner2, Thalita Faita2, Rodrigo M. Soares1, David A. V. da Silva1, Trícia
M. F. S. Oliveira2, Helena L. Ferreira2, Jaqueline A. Diniz1, Lara B. Keid2*

PT
RI
1
Departamento de Medicina Veterinária Preventiva e Saúde Animal, Faculdade de

SC
Medicina Veterinária e Zootecnia, Universidade de São Paulo, Pirassununga, Brasil
2
Departamento de Medicina Veterinária, Faculdade de Zootecnia e Engenharia de
NU
Alimentos, Universidade de São Paulo, Pirassununga, SP, Brasil
MA

* Corresponding author at: Av. Duque de Caxias Norte, 225, Jd. Elite, Pirassununga, SP,
Brasil, 13635-900.
D

Email address: [email protected] (L.B. Keid).

E
PT
CE
AC

1
 ACCEPTED MANUSCRIPT

ABSTRACT

Brucella canis, a gram-negative, facultative intracellular and zoonotic bacterium causes

canine brucellosis. Direct methods are the most appropriate for the detection of
canine brucellosis and bacterial isolation from blood samples has been employed as
gold-standard method. However, due to the delay in obtaining results and the
biological risk of the bacterial culturing, the polymerase chain reaction (PCR) has been
successfully used as an alternative method for the diagnosis of the infection. Sample
preparation is a key step for successful PCR and protocols that provide high DNA yield

PT
and purity are recommended to ensure high diagnostic sensitivity. The objective of this
study was to evaluate the performance of PCR for the diagnosis of B. canis infection in

RI
36 dogs by testing DNA of whole blood obtained through different extraction and
purification protocols. Methods 1 and 2 were based on a commercial kit, using

SC
protocols recommended for DNA purification of whole blood and tissue samples,
respectively. Method 3 was an in-house method based on enzymatic lysis and
purification using organic solvents. The results of the PCR on samples obtained through
NU
three different DNA extraction protocols were compared to the blood culture. Of the
36 dogs, 13 (36.1 %) were positive by blood culturing, while nine (25.0 %), 14 (38.8%),
and 15 (41.6%) were positive by PCR after DNA extraction using methods 1, 2 and 3,
MA

respectively. PCR performed on DNA purified by Method 2 was as efficient as blood

culturing and PCR performed on DNA purified with in-house method, but had the
advantage of being less laborious and, therefore, a suitable alternative for the direct B.
canis detection in dogs.
D

Keywords: Brucella canis, dogs, polymerase chain reaction, DNA purification.

E
PT
CE
AC

2
 ACCEPTED MANUSCRIPT

1. Introduction

Canine brucellosis is caused by Brucella canis, a gram-negative, facultative

intracellular and zoonotic bacterium that is associated with chronic and persistent
infections in dogs, with a worldwide distribution (Carmichael and Brunner, 1968;
CSFPH, 2012). Affected dogs may present with abortion, stillbirth, failure to conceive,
embryonic death, orchitis, epididymitis, prostatitis, enlargement of lymph nodes,

PT
discospondylitis, and uveitis (Keid et al., 2004; Carmichael and Greene, 2012). A
considerable proportion of infected dogs, including animals during the bacteraemic

RI
phase, may have no clinical manifestation but can act as sources of infection (Keid et
al., 2004).

SC
The control of canine brucellosis is hampered by the lack of rapid, accurate and
sensitive serological tests. Serological tests have limited value in canine brucellosis
NU
diagnosis because the humoral response may not be present, even during the
bacteraemic phase of infection (Keid et al., 2009, Wanke et al., 2012; Keid et al., 2015).
MA

Blood culture is the gold standard test for the diagnosis of infection since it
shows a higher sensitivity when compared to serological tests, along with a high
D

specificity, allowing the characterization of the isolated pathogen (Keid et al., 2007c;
E

Kim et al., 2007; Keid et al., 2009; Wanke et al., 2012; Keid et al., 2015). Indeed, during
PT

bacteremia the infected dogs usually eliminate high numbers of bacteria through
secretions and excretions (Keid et al., 2007 a,b) making the rapid identification of dogs
CE

during this phase a priority to reduce the transmission of the infection. The culturing
procedure, however, has the disadvantage of being time-consuming, prone to
AC

secondary contamination, and hazardous for laboratory personnel due to the zoonotic
nature of B. canis. In addition, a minimum amount of viable Brucella is required in
blood specimens for successful isolation (Alton et al., 1976; Keid et al., 2007c;
Carmichael and Greene, 2012).
Polymerase chain reaction (PCR) is a powerful tool for detecting Brucella
nucleic acids in various clinical specimens because it is less hazardous than culturing
(Wallach et al., 2004) and shows a higher sensitivity (Bricker, 2002; Kim et al., 2006;
Keid et al., 2007a,b,c; Aras and Uçan, 2010; Kang et al., 2014; Kauffman et al., 2014).
For these reasons, PCR has been successfully used as an alternative to blood culturing.

3
 ACCEPTED MANUSCRIPT

A large variety of methods have been developed for nucleic acid purification,
and those based on commercial kits are usually considered less laborious, less time-
consuming and more prone to standardization when compared to in-house methods
such as the standard phenol/chloroform protocol (Radström et al., 2004; Demeke and
Jenkins, 2010; Yuan et al., 2012). DNA purification kits based on the use of adsorption
of DNA to silica columns are among the most commercialized methods for purifying
DNA and they have proved to be efficient for a variety of samples; besides they are

PT
prone to automation (Hanselle et al., 2003; Rohland and Hofreiter, 2007; Demeke and
Jenkins, 2010; Tatcher, 2015). When several silica-based kits were used for DNA

RI
extraction of B. neotomae in milk and cheese (Lusk et al., 2013) and B. melitensis in
tissue samples (Tomaso et al., 2010), the Qiagen kits provided highest DNA yield when

SC
compared to the other kits.
Few studies have been described aiming B. canis detection in dogs using
NU
molecular assays (Kim et al., 2006; Keid et al., 2007a,b,c; Aras and Uçan, 2010; Kang et
al., 2014; Kauffman et al., 2014). From the published studies on B. canis detection,
MA

two methods for DNA extraction were employed: the proteinase K/phenol/chloroform
method (Keid et al., 2007a,b,c; Noosud et al., 2009) and Qiagen DNA column
D

purification kit (Kang et al., 2014; Kauffman et al., 2014). From the results obtained, it
E

seems that higher values of sensitivity were achieved by the proteinase

PT

K/phenol/chloroform method. However, those studies did not aim to evaluate nucleic
acid extraction methods. In addition, they employed different primers and
CE

amplification conditions hampering proper comparison between the protocols.

Mechanical and enzymatic-based pre-lysis steps prior to DNA extraction have
AC

been proved to enhance the concentration of bacterial DNA (Yuan et al., 2012). Thus,
besides the method for DNA extraction, the sample preparation is a key factor for
successful PCR. Protocols that can provide the highest DNA yield and purity, as well as
reproducibility, are recommended to ensure high diagnostic sensitivity of the
forthcoming DNA-based diagnostic test.
In view of the usefulness of PCR for the canine brucellosis diagnosis,
improvements in protocol for DNA extraction are of utmost importance to augment
the sensitivity and rapidity of the test. Therefore, the objective of the present study
was to compare an artisanal well-known and efficient protocol for DNA extraction with

4
 ACCEPTED MANUSCRIPT

a widely used commercial kit as well as to evaluate the efficiency of protocols for B.
canis DNA extraction in canine blood samples for the first time.

2. Materials and methods

2.1. Animals and samples

PT
Forty dogs were used in this study. Of these, 36 are referred in this study as
testing dogs and include 27 dogs from a kennel where episodes of abortion occurred

RI
due to a B. canis outbreak; and nine pet dogs that were not used for reproduction
purposes that presented with no clinical or epidemiological evidence of brucellosis. In

SC
addition, four dogs from a kennel that was free of brucellosis were used to determine
the analytical sensitivity of the molecular assays. These four dogs came from a kennel
NU
that had been monitored with serological and microbiological tests with negative
results in the last four years and are referred to in this study as brucellosis-free dogs.
MA

From each dog, a total of 5-8 ml of blood was aseptically collected through
jugular or brachial venipuncture and stored in tubes containing sodium citrate as the
D

anticoagulant. Samples were stored at 4◦C during sampling and were laboratory-
E

processed after a maximum of 12 h.

PT

2.2. Blood culturing

CE

Blood samples from the 40 dogs were submitted for blood culturing according
AC

to previously described protocols (Alton et al., 1976; Keid et al., 2007c). Isolated
bacterial colonies were identified by macroscopic morphology, Gram stain, and Bruce-
ladder assay (López-Goñi et al., 2011).

2.3. Brucella canis RM6/66 culturing

Brucella canis reference strain RM6/66 (ATCC 23365) was used in this study.
Cultures were reactivated from frozen stocks kept at -80◦C and streaked onto Tryptose
agar plates with 5% of fetal calf serum (FCS). Plates were incubated for 72 h at 37◦C

5
 ACCEPTED MANUSCRIPT

under an aerobic atmosphere. After bacterial growth, a single colony was transferred
to a tube containing 1 ml of sterile 0.9 % sodium chloride (NaCl) solution and
homogenized using a vortex at low speed for 15 s. An aliquot of 100 µl of the
suspension was streaked on the Tryptose agar plate with 5% FCS and incubated for 72
h at 37◦C under an aerobic atmosphere. After a purity confirmation by Gram staining
and an evaluation of the biochemical profile (Alton et al., 1976), cultures were
transferred to a microtube containing 1 ml of sterile 0.9 % NaCl solution, and the

PT
suspensions were stored at -20◦C until use for DNA extraction.

RI
2.4. Preparing serial dilutions of the B. canis RM6/66 strain in TE buffer

SC
An aliquot of 200 µl of the B. canis RM6/66 bacterial suspension was mixed
with DNA lysis buffer (10 mM Tris-HCl, 25 mM EDTA, 1 % SDS, 0.8 mg/mL proteinase K)
NU
to a final volume of 500 µl and incubated at 37◦C overnight.
Purification of the lysate was conducted using 5 mM NaCl and 1%
MA

hexadecyltrimethylammonium bromide (CTAB)/70 mM NaCl solution, with incubation

at 65◦C for 10 min to remove cell wall debris and polysaccharides (Ausubel et al.,
D

1999), followed by phenol:chloroform extraction and alcohol precipitation according

E

to Sambrook et al. (1989). The final DNA sediment was resuspended in 30 µl TE buffer
PT

(10 mM Tris-HCl, pH 8.0, and 1 mM EDTA, pH 8.0).

DNA concentration and quality were estimated spectrophotometrically (DS-11
CE

FX, DeNovix Inc., USA) by reading the optical densities at 260 and 280 wavelengths. B.
canis DNA obtained at a concentration of 2.12 ng/µL (260/280 ratio: 1.8; 260 reading:
AC

0.425) was tenfold diluted in TE buffer to obtain ten suspensions with the following
DNA concentrations: 212.60 ng/µl, 21.26 ng/µl, 2.12 ng/µl, 212.60 pg/µl, 21.26 pg/µl,
2.12 pg/µl, 212.60 fg/µl, 21.26 fg/µl, 2.12 fg/µl, and 0.21 fg/µl. The obtained
suspensions were stored at -20◦C until use as the template in the PCR assays.

2.5. Whole-blood pre-lysis testing

To enhance the recovery of B. canis DNA from blood samples of naturally

infected dogs, three protocols for whole-blood pre-lysis prior to DNA extraction were

6
 ACCEPTED MANUSCRIPT

compared. Blood samples collected from four testing dogs that had positive results by
blood culturing were used. From each dog, four aliquots of 1 ml of blood were thawed
and repeatedly washed according to Keid et al. (2007c), until the leukocyte sediment
lost its reddish colour. Each pellet obtained from each dog was lysed using the four
methods mentioned below.
Method A used an enzymatic pre-lysis. The leukocyte sediment was
resuspended in a buffer containing 20 mM Tris-HCl, 2 mM EDTA, 20 mg/ml lysozyme,

PT
and 1 % Triton X-100. Samples were mixed and incubated at 37◦C for 2 h. Afterwards,
samples were centrifuged at 13,000 x g for 1 min and mixed with 500 µl of lysis buffer

RI
(10 mM Tris-HCl, 25 mM EDTA, 1 % SDS, 0.8 mg/mL proteinase K), homogenized, and
incubated at 37◦C overnight.

SC
Method B used a mechanical pre-lysis. The sediment was resuspended directly
in 500 µl of the lysis buffer (10 mM Tris-HCl, 25 mM EDTA, 1 % SDS, 0.8 mg/mL
NU
proteinase K) with a mixture of 1.0 mm zirconia and 0.1 mm of silica/zirconia beads
(BioSpec, Bartlesville, OK). Samples were incubated at 37◦C overnight and
MA

homogenized on a full-speed vortex for 1 min every 15 min, six times.

Method C comprised the use of pre-lysis protocols A and B together.
D

In method D, the blood sediment of each dog was resuspended directly in lysis
E

buffer without pre-lysis treatment.

PT

After pre-lysis, all samples were submitted to the same DNA lysis and
purification procedure described in item 2.4 (Sambrook et al., 2001; Ausubel et al.,
CE

1999). For the pre-lysis methods using beads (B and C), the lysates were briefly
centrifuged and transferred (except beads) to a new tube before following the
AC

purification procedure.
The pre-lysis treatment that revealed the strongest PCR band intensity upon
visual examination of gel electrophoresis after PCR was included in each protocol for
whole blood DNA extraction described in item 2.6. (Figure 1). Electrophoresis was
conducted in a 1.5% agarose gel, followed by ethidium bromide (25 µg/ml) staining for
1 h, according to Sambrook et al. (2001).

2.6. Whole-blood DNA extraction

7
 ACCEPTED MANUSCRIPT

Three DNA extraction methods were performed on the 40 canine blood

samples (Figure 1). The three methods for DNA extraction were conducted as follows:
(i) Extraction method 1 used the protocol for isolation of DNA from whole non-
nucleated animal blood of the DNeasy Blood and Tissue kit (Qiagen, Hilden, Germany)
to extract DNA from 100 µl of each canine whole blood sample. After pre-lysis, a lysis
buffer containing 200 µl of AL buffer and 20 µl of proteinase K was added, and samples
were homogenized, incubated at 56◦C overnight, and processed on a full-speed vortex

PT
for 1 min, every 15 min, six times. Next, the mixture was centrifuged at full speed
(13,000 x g at room temperature) for 1 min and the top aqueous phase was

RI
transferred to a new microtube. Purification was conducted according to the
manufacturer’s instructions, and the obtained DNA was eluted in 50 µl of elution

SC
buffer (AE buffer).
(ii) Extraction method 2 used the leukocyte sediment obtained from 1 ml of blood
NU
according to the washing procedure described in item 2.5. After pre-lysis, samples
were extracted using the protocol for isolation of total DNA from animal tissues of the
MA

DNeasy Blood and Tissue kit (Qiagen, Hilden, Germany), using 180 µl of ATL buffer and
20 µl of proteinase K. Samples were then homogenized, incubated at 56◦C overnight,
D

and processed on a full-speed vortex for 1 min every 15 min, six times. The mixture
E

was centrifuged at full speed (13,000 x g at room temperature) for 1 min and the
PT

aqueous layer was transferred to a new microcentrifuge tube, and purification was
performed according to the manufacturers’ instructions. The obtained DNA was eluted
CE

in 50 µl of AE buffer.
(iii) Extraction method 3 used 1 ml of each blood sample, which was washed exactly as
AC

described in item 2.5 for haemoglobin removal. After pre-lysis, lysis buffer containing
10 mM Tris-HCl, 25 mM EDTA, 1 % SDS, and 0.8 mg/mL proteinase K was added to a
final volume of 500 µl. Samples were homogenized, incubated at 37◦C overnight, and
processed on a full-speed vortex for 1 min every 15 min, six times. After centrifugation
at full speed (13,000 x g at room temperature) for 1 min the aqueous layer was
transferred to a new microcentrifuge tube and purified exactly as described for DNA
extraction of the B. canis strain, using CTAB extraction (Ausubel et al., 1999),
phenol/chloroform purification, and alcohol precipitation (Sambrook et al., 2001). The
obtained DNA was eluted in 30 µl of TE buffer.

8
 ACCEPTED MANUSCRIPT

2.7. Spiking of canine DNA obtained from whole blood of brucellosis-free dogs with
decreasing amounts of B. canis DNA

DNA samples obtained from the blood of the four brucellosis-free dogs
according to the three extraction methods described in section 2.6 were quantified
spectrophotometrically (DS-11 FX, DeNovix Inc., USA) by reading the optical densities

PT
at 260 and 280 wavelengths, for a total of 12 DNA samples comprising four samples
per extraction method.

RI
Then, the four DNA samples obtained through each extraction protocol were
pooled and homogenized. B. canis DNA obtained at a concentration of 2.12 ng/µL (see

SC
item 2.4) was diluted tenfold in each one of the four pools of canine DNA to obtain ten
suspensions of canine DNA spiked with the following B. canis DNA concentrations:
NU
212.60 ng/µl, 21.26 ng/µl, 2.12 ng/µl, 212.60 pg/µl, 21.26 pg/µl, 2.12 pg/µl, 212.60
fg/µl, 21.26 fg/µl, 2.12 fg/µl, and 0.212 fg/µl.
MA

2.8. Polymerase chain reaction (PCR)

E D

PCR was conducted on DNA obtained from the following samples: whole blood
PT

of the 36 testing dogs through the three different extraction protocols; B. canis that
was extracted and tenfold diluted in TE buffer according to section 2.4; and whole
CE

blood of the four brucellosis-free dogs that were pooled and spiked with decreasing
concentrations of B. canis DNA (see section 2.7).
AC

PCR was conducted using a pair of primers directed to the 16S-23S rRNA
interspace region of Brucella spp., as described by Keid et al. (2007). In addition, PCR
assays targeting a canine housekeeping gene (the β-actin gene) were performed in all
canine DNA samples that yielded negative results in the Brucella-specific PCR analysis,
as an internal control for the reaction, as previously described (Greer et al., 1991).
To avoid cross-contamination of samples, a negative control (ultrapure water)
was included for every two samples throughout the molecular procedures. In addition,
the laboratory was divided into separated areas for sample preparation, DNA
extraction, preparation of the amplification mixture, PCR amplification, and

9
 ACCEPTED MANUSCRIPT

electrophoresis. Each laboratory area was equipped with anti-contamination

equipment such as filter tips, a DNA workstation, ultraviolet light for sterilization of
surfaces, individual laboratory coats, and gloves. All surfaces were cleaned daily with
2% sodium hypochlorite.

2.9. Statistical analysis

PT
The pairwise comparisons of the results of blood culturing and PCR using
different DNA extraction protocols were carried out using McNemar’s test and Kappa

RI
coefficient (95 % confidence interval) (McNemar et al., 1947; Altman, 1991; Bland,
2000; Cohen, 1960).

3. Results
SC
NU
The four samples of blood from the dogs with positive results through blood
MA

culturing, which were used to assess the efficiency of the pre-lysis treatment, yielded
positive PCR results. However, samples treated with the combination of enzymatic and
D

mechanical pre-lysis (method C) yielded stronger amplicons. Therefore, the combined

E

pre-lysis protocol was included prior to DNA extraction of the 36 whole blood samples
PT

that was conducted using extraction methods 1, 2 and 3.

The analytical sensitivity of PCR using decreasing amounts of B. canis DNA
CE

resuspended in TE buffer was 21.26 fg/µL. PCR using pooled DNA extracted from the
blood of the four brucellosis-free dogs using the three different extraction protocols
AC

and spiked with decreasing amounts of B. canis DNA also allowed the detection of
21.26 fg/µL of B. canis DNA, regardless of the DNA extraction method used. The DNA
concentrations of the blood samples from the brucellosis-free dogs that were
determined using the three different extraction methods are shown in Table 2. The
average concentrations of DNA obtained through method 3 were almost twice the
values obtained through methods 1 and 2.
Of the 36 testing dogs, 13 (36.1 %) yielded positive results through blood
culturing, thus confirming Brucella infection. All the isolates were characterized as B.
canis in Bruce-ladder assay. All positive samples were obtained from dogs from a

10
 ACCEPTED MANUSCRIPT

commercial kennel with a history of abortion, stillbirths, and failure to conceive. All
samples obtained from the nine pet dogs were negative by blood culturing.
Of the 36 blood samples, 9 (25 %), 14 (38.8%), and 15 (41.6%) yielded positive
results by PCR after DNA extraction using methods 1, 2 and 3, respectively. Sixteen of
the 36 dogs (44.4 %) had positive results in at least one test; all were from the
commercial kennel.
All dogs with negative results by Brucella PCR yielded positive results in β-actin

PT
PCR, indicating absence of amplification inhibitors.
The results of the agreement analysis between each diagnostic method pair-

RI
compared can be observed on Table 1. A perfect agreement was observed between
blood culturing and PCR using extraction method 3, while the agreement between

SC
blood culturing and methods 1 and 2 was substantial. The agreement between
extraction methods 2 and 3 was considered almost perfect.
NU
4. Discussion and conclusions
MA

Brucella is a gram-negative, facultative intracellular bacterium found mainly in

D

monocytes during the bacteraemic phase of canine brucellosis, so efficient cell lysis
E

procedures must be employed to enable high amounts of target DNA. The extraction
PT

protocol needs to guarantee leucocyte lysis followed by efficient rupture of the

bacterial cell wall. Most studies regarding the evaluation of pre-lysis procedures for
CE

DNA extraction relies on the parallel extraction of samples with and without pre-lysis
step, thus comparing the DNA yield visually through the intensity of gel staining
AC

(Yeates et al., 1998; Kay et al., 2011) or by determining DNA concentration, through
quantitative PCR assay or spectrophotometrically (Burke et al., 2016).
In the present study, the efficiency of the pre-lysis methods in extracting DNA
from bacteria during intracellular life in naturally infected dogs was determined
visually according to the intensity of the amplicons. It was assumed that the method
able to produce brighter bands had the potential to enhance the yield of bacterial DNA
and thus the sensitivity of the assay, probably as consequence of the improvement in
Brucella cell lysis (Yuan et al., 2012). It therefore should be useful for samples with low
bacterial numbers. Thus, the combined pre-lysis protocol was included before each

11
 ACCEPTED MANUSCRIPT

extraction method. The use of lysozyme is known to allow the efficient extraction of
DNA from gram-positive bacteria that have cell walls with thick layers of peptidoglycan
(Phillips, 1966). Although gram-negative bacteria have a lower proportion of
peptidoglycan in the cell wall, the use of lysozyme associated with Triton X-100 may
enhance cell wall lysis of gram-negative bacterial cell wall, as well (Tatcher, 2015).
In view of the promising results obtained by using the combined pre-lysis step
in the four blood samples analysed through gel visualization, further studies should be

PT
conducted in the future to achieve a more precise evaluation of pre-lysis strategies for
intracellular bacteria using samples from naturally infected hosts. The determination

RI
of the average number of bacteria in white blood cell using flow cytometry assay
(Käser et al., 2016), and correlating it with DNA yield and positivity should be useful for

SC
this intent.
In the present study, caution was taken to not include NaCl in the lysis buffer to
NU
avoid the binding of extracted DNA to the silica beads used for mechanical disruption.
It is well known that the presence of divalent and monovalent cations, such as Na+,
MA

can provide conditions for DNA adsorption to the silica surface (Rohland and Hofreiter,
2007; Carlson et al., 2014). Since NaCl is important for DNA precipitation, it was
D

included in the purification step, after the lysis step and the transfer of the lysates to
E

new microtubes without silica beads before starting the purification procedure.
PT

No difference was observed in the analytical sensitivity of the PCR assay using
decreasing amounts of B. canis DNA diluted in TE buffer or in the presence of canine
CE

DNA obtained from whole blood through different extraction protocols. Therefore, the
three DNA extraction methods proved to be equally efficient towards removing
AC

inhibitors from the blood samples, such as haemoglobin, its derivative hemin present
in the erythrocyte, and lactoferrin present in the leukocyte (Akane et al., 1994; Abu Al-
Soud, et al., 2010). In methods 2 and 3, where more whole blood was used for
extraction, a washing step was employed for osmotic lysis of erythrocytes prior to cell
lysis to guarantee adequate removal of large amounts of haemoglobin. In addition, all
samples with negative results through the Brucella PCR had positive results in the β-
actin PCR, another indicative of the absence of amplification inhibitors. It should be
noted that the approach used herein to determine the analytical sensitivity of the
assays does not allow a determination of the efficiency of the lysis methods. B. canis

12
 ACCEPTED MANUSCRIPT

DNA that was already extracted was used to spike the canine DNA obtained from the
blood, which was a more precise method to determine the detection threshold of the
assay. However, it does not consider the effect of bacterial cell wall or host cell on the
efficiency of the lysis.
The number of positive samples detected by extraction methods 2 and 3 was
significantly higher than that obtained by method 1 (Table 1). The PCR detection limit
was similar for the three methods, thereby indicating the absence of inhibitors,

PT
therefore, the differences in the number of positive samples detected among the 36
testing dogs may be a consequence of the different amounts of blood used for the

RI
DNA extraction. Higher amounts of blood were used for the DNA extraction in
methods 2 and 3 (1 ml) when compared to method 1 (100 µl), which could act by

SC
enhancing the amount of Brucella cells available for DNA extraction, thus leading to a
higher positivity. To avoid haemoglobin interference in the amplification reaction
NU
when using larger amounts of blood, a washing step was used to remove haemoglobin
prior to lysis. The washing step was possible because of the intracellular nature of
MA

Brucella, allowing the bacteria to be centrifuged inside the circulating leukocytes.

Another possible explanation should be the different composition of the lysis buffer
D

used in the three extraction methods. All methods employed the same pre-lysis step
E

and proteinase K in the lysis buffer, but only methods 2 and 3 used SDS in the lysis
PT

buffer, while in method 1, guanidine hydrochloride was used.

When comparing methods 2 and 3, the number of positive animals detected by
CE

PCR in samples extracted by method 3 was slightly higher than the number detected
by method 2. Both methods used an initial volume of blood of 1 ml for DNA extraction.
AC

Therefore, this slight difference should be explained by the higher amount of DNA
purified through method 3 when compared to method 2, enhancing the probability of
Brucella DNA detection (Table 2). In method 3, all extracted DNA was precipitated by
isopropanol, while in method 2, a smaller amount of DNA was expected to be obtained
since the yield was limited by saturation of the silica-gel membrane during the
purification step. In addition, the minimum recommended volume to elute DNA from
columns in the DNeasy Blood and Tissue kit was larger than the volume used to elute
DNA after the phenol/chloroform method (50 µl vs 30 µl).

13
 ACCEPTED MANUSCRIPT

According to our results, the protocol recommended by the DNeasy Blood and
Tissue Kit for DNA extraction of blood samples did not achieve an adequate sensitivity
to be used as an alternative for blood culturing for canine brucellosis diagnosis,
although it required less labour and less time to be conducted, because the blood
washing and DNA precipitation steps were not needed.
Otherwise, methods 2 and 3 yielded similar values of sensitivity and should be
used for canine brucellosis diagnosis in the field. Both methods can be completed in 2

PT
days, a period of time significantly lower than that required for blood culturing, but the
labour required in method 3 was higher when compared to method 2 because a DNA

RI
precipitation step is needed. In addition, method 2 does not require the use of harmful
chemicals such as phenol and chloroform. Therefore, the method based on the tissue

SC
protocol of DNeasy Blood and Tissue Kit proved to be a good alternative for direct B.
canis detection in dogs using blood samples.
NU
Acknowledgements
MA

Our thanks to FAPESP (Research Support Foundation of the State of São Paulo) for the

research support grant (process number 2015/06072-9); CNPq (National Council for
D

Scientific and Technological Development) for the doctorate fellowship granted to

E
PT

Lima JTR; Coordination for the Improvement of Higher Education Personnel (CAPES)

for the fellowship granted to Batinga MCA; and the University of São Paulo for the
CE

scholarship to dos Santos JC.

AC

References

Abu Al-Soud, W., Rådström, P., 2001. Purification and characterization of PCR
inhibitory components in blood cells. J. Clin. Microbiol. 39, 485–493

Akane, A., Matsubara, K., Nakamura, H., Takahashi, S., Kimura, K., 1994. Identification
of the heme compound copurified with deoxyribonucleic acid (DNA) from

14
 ACCEPTED MANUSCRIPT

bloodstains, a major inhibitor of polymerase chain reaction (PCR) amplification. J.

Forensic Sci. 39, 362–372.

Alton, G.G., Jones, L.M., Angus, R.D., Verges, J.M., 1976. Las Tecnicas de Laboratorios
en la Brucelosis, 2nd ed. Ginebra: Organización Mundial de la Salud, 109pp.

Altman, D.G., 1991. Practical statistics for medical research. London: Chapman and
Hall, 202 pp.

PT
Aras, Z., Uçan, U.S., 2010. Detection of Brucella canis from inguinal lymph nodes of

RI
naturally infected dogs by PCR. Theriogenol. 74:658–662.

SC
Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., Struhl,
K., editor. 1999. Short Protocols in Molecular Biology. New York: Wiley & Sons,
NU
1512p.

Bland, M., 2000. An introduction to medical statistics, 3rd ed. Oxford: Oxford
MA

University Press.

Bricker, B.J., 2002. PCR as a diagnostic tool for brucellosis. Vet Microbiol, 90:435-446.
E D

Burke, R.M., McKenna, J.P., Cox, C., Coyle, P.V., Shields, M.D., Fairley, D.J., 2016. A
PT

comparison of different pre-lysis methods and extraction kits for recovery of

Streptococcus agalacticae (Lancefield group B Streptococcus) DNA from whole
CE

blood. J. Microbiol. Methd. 129:103-108.

AC

Carlson, K., Flick, L., Hall, M., 2014. DNA adsorption onto calcium aluminate and silicate
glass surfaces. Colloids Surf B Biointerfaces 117:538–544.

Carmichael, L.E., Bruner, D.W., 1968. Characteristics of a newly-recognized species of

Brucella responsible for infectious canine abortions. Cornell Vet. 48:579-592, 1968.

Carmichael, L.E, Greene, C.E., 2012. Canine brucellosis. In: Greene, C.E. Infectious
diseases of dog and cat, 4. ed. Missouri: Saunders, p. 398-411.

15
 ACCEPTED MANUSCRIPT

Cohen J. A., 1960. Coefficient of agreement for nominal scales. Educ Psychol Meas.
20:37–46.

CSFPH, 2012. Canine Brucellosis: Brucella canis. 2012: http://www.c

fsph.iastate.edu/Factsheets/pdfs/brucellosis_canis.pdf

Demeke, T., Jenkins, G.R., 2010. Influence of DNA extraction methods, PCR inhibitors
and quantification methods on real-time PCR assay of biotechnology-derived traits.

PT
Anal. Bioanal. Chem. 396: 1977-1990.

RI
Greer, C.E., Peterson, S.L., Kiviat, N.B., Mamos, M.M., 1991. PCR amplification from
paraffin-embedded tissues. Effects of fixative and fixation time. Am. J. Clin. Pathol.

SC
95: 117-124. NU
Hanselle, T., Otte, M., Schnibbe, T., Smythe, E., Krieg-Schneider, F., 2003. Isolation of
genomic DNA from buccal swabs for forensic analysis, using fully automated silica-
MA

membrane purification technology, Leg. Med. (Tokyo) 5 (Suppl 1):S145–S149.

Kang, S.I., Lee, S.E., Kim, J.Y., Lee, K., Kim, J.W., Lee, H.K., Sung, S.R., Heo, Y.R., Jung,
D

S.C., Her, M., 2014. A new Brucella canis species-specific PCR assay for the diagnosis
E

of canine brucellosis. Comp. Immumol. Microbiol. Infect. Dis. 37:237-241.

PT

Käser, T., Pasternak, J.A.; Hamonic, G., Rieder, M., Lai, K., Delgado-Ortega, M., Gerdts,
V., Meurens, F., 2016. Flow Cytometry as an Improved Method for the Titration of
CE

Chlamydiaceae and Other Intracellular Bacteria. Cytometry Part A 89A: 451-460.

AC

Kauffman, L.K., Bjork, J.K., Gallup, J.M., Boggiatto, P.M., Bellaire, B.H., Petersen, C.A.,
2014. Early detection of Brucella canis via quantitative polymerase chain reaction
analysis. Zoon. Publ. Health 61:48-54.

Kay, M.K., Link, L., Triantis, J., Salman, M.D., Larssen, R.S., 2011. Evaluation of DNA
extraction techniques detecting Mycobacterium tuberculosis complex organisms in
Asian elephant trunk wash samples. J. Clin. Microbiol. 49:618-623.

16
 ACCEPTED MANUSCRIPT

Keid, L.B., Soares, R.M., Morais, Z.M., Richtzenhain, L.J., Vasconcellos, S.A., 2004.
Brucella spp. isolation from dogs from commercial breeding kennels in São Paulo,
Brazil. Braz. J. Microbiol. 35:161-166.

Keid, L.B., Soares, R.M., Vasconcellos, S.A., Chiebao, D.P., Salgado, V.R., Megid, J.,
Richtzenhain, L.J., 2007a. A polymerase chain reaction for detection of Brucella
canis in vaginal swabs of naturally infected bitches. Theriogenol. 68:1260-1270.

PT
Keid, L.B., Soares, R.M., Vasconcellos, S.A., Chiebao, D.P., Megid, J., Salgado, V.R.,
Richtzenhain, L.J., 2007b. A polymerase chain reaction for the detection of Brucella

RI
canis in semen of naturally infected dogs. Theriogenol. 67:1203-1210.

SC
Keid, L.B., Soares, R.M., Vieira, N.R., Megid, J., Salgado, V.R., Vasconcellos, S.A., Costa,
M., Gregori, F., Richtzenhain, L.J., 2007c. Diagnosis of canine brucellosis:
NU
comparison between serological and microbiological tests and a PCR based on
primers to 16S-23S rDNA interspacer. Vet. Res.Commun. 31:951-965.
MA

Keid, L.B., Soares, R.M., Vasconcellos, S.A., Megid, J., Salgado, V.R., Richtzenhain, L.J.,
2009. Comparison of agar gel immunodiffusion test, rapid slide agglutination test,
D

microbiological culture and PCR for the diagnosis of canine brucellosis. Res. Vet. Sci.
E

86:22-26.
PT

Keid, L.B., Diniz, J.A., Oliveira, T.M.F.S., Ferreira, H.L., Soares, R.M., 2015. Evaluation of
CE

an immunochromatograhic test to the diagnosis of canine brucellosis caused by

Brucella canis. Reprod. Dom. Anim. 50:939-944.
AC

Kim, J.W., Lee, V.J., Han, M.Y., BAE, D.H., Jung, S.C., OH, J.S., Ha, G.W., Cho, B.K., 2007.
Evaluation of immunochromatographic assay for serodiagnosis of Brucella canis. J.
Vet. Med Sci. 69:1103-1107.

Kim, S., Lee, D.S., Suzuki, H., Watarai, M., 2006. Detection of Brucella canis and
Leptospira interrogans in canine semen by multiplex nested PCR. J. Vet. Med. Sci.
68: 615–618.

17
 ACCEPTED MANUSCRIPT

López-Goñi, I., García-Yoldi, D., Marín, C.M., de Miguel, M.J., Barquero-Calvo, E.,
Guzmán-Verri, C., Albert, D., Garin-Bastuji, B., 2011. New Bruce-ladder multiplex
PCR assay for the biovar typing of Brucella suis and the discrimination of Brucella
suis and Brucella canis. Vet. Microbiol. 154:152-155.

Lusk, T.S., Strain, E., Kase, J.A., 2013. Comparison of six commercial DNA extraction kits
for detection of Brucella neotomae in Mexican and Central American-style cheese
and other milk products. Food Microbiol. 34:100-105.

PT
McNemar, Q., 1947. Note on the sampling error of the difference between correlated

RI
proportions or percentages. Psychom. 12:153-157.

SC
Noosud, J., Sirinarumitr, K., Sirinarumitr, T., 2009. Comparison of a Serological
Method, a Bacteriological Method and 16S rRNA Brucella canis PCR for Canine
NU
Brucellosis Diagnosis. Kasetsart J. (Nat. Sci.) 43:159-164.
MA

Phillips, D.C., 1966. The three-dimensional structure of an enzyme molecule. Sci Am

215: 78–90.
D

Rådström, P., Knutsson, R., Wolffs, P., Lövenklev, M., Löfström, C., 2004. Pre-
E

PCR processing: strategies to generate PCR-compatible samples. Mol. Biotechnol.

PT

26:133-46.

Rohland, N., Hofreiter, M., 2007. Comparison and optimization of ancient DNA
CE

extraction. BioTechniques 42:343–352.

AC

Sambrook, J., Russell, D.W., 2001. Molecular cloning: a laboratory manual, 3 rd ed.
New York: Cold Spring Harbor Laboratory Press, 999p.

Tatcher, S.A., 2015. DNA/RNA Preparation for Molecular Detection. Clin. Chem. 61:89-
99.

Tomaso, H., Kattar, M., Eickhoff, M., Wernery, U., Al Dahouk, S., Straube, E., Neubauer,
H., Scholz, H.C, 2010. Comparison of of commercial DNA preparation kits for the

18
 ACCEPTED MANUSCRIPT

detection of Brucellae in tissue using quantitative real-time PCR. BMC Infect. Dis.
10:100.

Yeates, C., Gillings, M.R., Davison, A.D., Altavilla, N., Veal, D.A., 1998. Methods for
microbial DNA extraction from soil for PCR amplification. Biol. Proced. Online 1:40-
47.

Yuan, S., Cohen, D.B., Ravel, J., Abdo, Z., Forney, L.J., 2012. Evaluation of Methods for

PT
the Extraction and Purification of DNA from the Human Microbiome. PlosOne 7:
(e33865).

RI
Wallach, J.C., Giambartolomei, G.H., Baldi, P.C., Fossati, C.A., 2004. Human Infection

SC
with M- Strain of Brucella canis. Emerg. Infect. Dis. 10:146-148.
NU
Wanke, M.M., Cairó, F., Rossano, M., Laiño, M., Baldi, P.C., Monachesi, M.E., Comercio,
E.A., Vivot, M.M., 2012. Preliminary study of an immunochromatography test for
MA

serological diagnosis of canine brucellosis. Reprod. Domest. Anim. 47:370-372.

E D
PT
CE
AC

19
 ACCEPTED MANUSCRIPT

PT
RI
SC
Figure 1
NU
MA
DE
PT
CE
AC

20
 ACCEPTED MANUSCRIPT

Table 1

Comparison of the results obtained through blood culture and polymerase chain reaction (PCR) to the detection of Brucella canis DNA in 36 canine whole
blood samples obtained by different extraction methods, using Mc Nemar test and Kappa coefficient.

Results Kappa

P T Mc Nemar Test

Test 1 Test 2
Positive by
test 1 and
Negative by Positive by Negative by
test 1 and test 1 and test 1 and
Value
R I
Agreement X 2
p value
Similar
proportion of
positive by
test 2
negative by negative
test 2 by test 2
positive by
test 2
S C positives

Blood culture
Extraction
method 1
9 23 4 0
N U
0.74 substantial 4.0 0.04 no

Blood culture
Extraction
method 2
11 20 2
M A
3 0.70 substantial 0.2 0.65 yes

Blood culture
Extraction
method 3
13 21
E D 0 2 0.88 almost perfect 2.0 0.15 yes

Extraction
method 1
Extraction
method 2
8
P T
21 1 6 0.56 moderate 3.57 0.05 yes

Extraction Extraction
C
9
E 21 0 6 0.63 substantial 6.0 0.01 no
method 1

Extraction
method 3

Extraction
A C13 20 1 2 0.82 almost perfect 0.33 0.56 yes
method 2 method 3

21
 ACCEPTED MANUSCRIPT

Table 2: DNA concentration of blood samples from brucellosis free dogs obtained though
different extractions methods.

260/280
Sample Extraction Method DNA Concentration (ng/uL) A260
Ratio
Brucellosis free dog 1 1 49.61 0.99 1.67
Brucellosis free dog 2 1 19.20 0.38 1.70
Brucellosis free dog 3 1 18.37 0.36 1.64
Brucellosis free dog 4 1 8.67 0.17 2.05
Average 23.96
Brucellosis free dog 1 2 18.25 0.36 2.20

PT
Brucellosis free dog 2 2 32.73 0.65 1.69
Brucellosis free dog 3 2 28.64 0.57 1.90
Brucellosis free dog 4 2 10.23 0.20 2.60

RI
Average 22,46
Brucellosis free dog 1 3 38.19 0.76 1.65

SC
Brucellosis free dog 2 3 51.20 1.02 1.65
Brucellosis free dog 3 3 30.75 0.61 1.87
Brucellosis free dog 4 3 53.34 1.06 1.82
NU
Average 43,37
MA
E D
PT
CE
AC

22
 ACCEPTED MANUSCRIPT

Highlights

The use of a pre-lysis step improved DNA extraction of Brucella canis cells.

The three DNA extraction methods yielded the same analytical sensitivity.

Differences in the number of positive dogs detected were observed depending on the

method us

PT
RI
SC
NU
MA
E D
PT
CE
AC

23