Development of a rapid and visual nucleotide detection method for a Chinese epidemic strain of Orientia tsutsugamushi based on recombinase polymerase amplification assay and lateral flow test

Accepted Manuscript Title: Development of a rapid and visual nucleotide detection method towards an Chinese epidemic strain of Orientia tsutsugamushi based on recombinase polymerase amplification assay and lateral flow test Authors: Yong Qi, Qiong Yin, Yinxiu Shao, Min Cao, Suqin Li, Hongxia Chen, Wanpeng Shen, Jixian Rao, Jiameng Li, Xiaoling Li, Yu Sun, Yu Lin, Yi Deng, Wenwen Zeng, Shulong Zheng, Suyun Liu, Yuexi Li PII: DOI: Reference:

S1201-9712(18)30062-6 https://doi.org/10.1016/j.ijid.2018.03.003 IJID 3190

To appear in:

International Journal of Infectious Diseases

Received date: Revised date: Accepted date:

19-1-2018 19-2-2018 8-3-2018

Please cite this article as: Qi Yong, Yin Qiong, Shao Yinxiu, Cao Min, Li Suqin, Chen Hongxia, Shen Wanpeng, Rao Jixian, Li Jiameng, Li Xiaoling, Sun Yu, Lin Yu, Deng Yi, Zeng Wenwen, Zheng Shulong, Liu Suyun, Li Yuexi.Development of a rapid and visual nucleotide detection method towards an Chinese epidemic strain of Orientia tsutsugamushi based on recombinase polymerase amplification assay and lateral flow test.International Journal of Infectious Diseases https://doi.org/10.1016/j.ijid.2018.03.003 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.

Development of a rapid and visual nucleotide detection method towards an Chinese epidemic strain of Orientia tsutsugamushi based on recombinase

IP T

polymerase amplification assay and lateral flow test

Yong Qi1，#, Qiong Yin2，#, Yinxiu Shao2, Min Cao1, Suqin Li1, Hongxia Chen1,

SC R

Wanpeng Shen1, Jixian Rao1, Jiameng Li1, Xiaoling Li2, Yu Sun3, Yu Lin2, Yi Deng2,

N

U

Wenwen Zeng2, Shulong Zheng2, Suyun Liu2, Yuexi Li1，2， 3*

M

Jiangsu Province, China,

A

1 Huadong Research Institute for Medicine and Biotechniques, Nanjing 210002,

ED

2 China Pharmaceutical University, Nanjing 210002, Jiangsu Province, China

PT

3 Nanjing Medical University, Nanjing 210002, Jiangsu Province, China

CC E

# These authors contributed equally to this research

A

* Corresponding author, [email protected]

1

Highlights Diagnosing scrub typhus is a challenge and modest diagnostic tests is in need



Visually diagnostic test suitable for use in basic medical units was developed



The detection limits are 10 copies/reaction



The sensitivity and specificity are comparable to those of quantitative PCR



The detection can be completed within 30 min at 37oC

SC R

IP T



U

Abstract

N

Objectives Orientia tsutsugamushi is an obligate intracellular pathogen that

A

causes scrub typhus. Diagnosing scrub typhus is still a challenge and sensitive, specific,

M

simple, and rapid diagnostic tests is in need.

ED

Methods Recombinase polymerase amplification (RPA) assay combined with a lateral flow (LF) test was developed and optimized targeting 56-kDa gene of a

PT

Karp-like strain of O. tsutsugamushi. Detection limits, sensitivity, specificity, and

CC E

simulative clinical performance were evaluated. Results Primers and probe were screened to establish the RPA assay and the

A

reaction conditions were optimized. The detection limits were 10 and 12 copies/reaction in detecting plasmid and genomic DNA, respectively. The RPA-LF method could differentiate O. tsutsugamushi from other phylogenetically related bacteria. The sensitivity and specificity were 100% and over 90%, respectively, 2

evaluated using infected animal samples or simulative clinical samples. Furthermore, the method was completed in 20 min at 37oC followed by a 3 to 5 min incubation at room temperature for development of an immunochromatographic strip and results can be determined visually.

IP T

Conclusions, The method is promising for wide-ranging use in basic medical

SC R

units considering its low requirement of instruments and infrastructure as well as highly time-efficient, sensitive and specific for diagnosing scrub typhus.

Key words: Orientia tsutsugamushi, scrub typhus, recombinase polymerase

N

U

amplification, diagnosis, lateral flow test

A

Introduction

M

Orientia tsutsugamushi is an obligate intracellular pathogen that causes scrub

ED

typhus, an acute febrile illness. O. tsutsugamushi is transmitted through the bite of chiggers, commonly encountered in rural areas. Scrub typhus was a major concern in

PT

the preantibiotic era and caused thousands of fatalities during the Second World War,

CC E

with the case fatality rate approaching 60% (Beran and Steele, 1994). This disease is endemic to a region known as the ‘tsutsugamushi triangle’, extending from Japan and

A

Russia in the North to Northern Australia in the South and the Arabian peninsula in the West (Janardhanan et al., 2014, Xu et al., 2017), with around 1 million cases of this disease occur annually and 1 billion people are at risk of infection (Rosenberg, 1997, Xu et al., 2017). Within this area, China is an important epidemic area with its broad 3

territorial area, large population, and prevalence of ecotourism. Nowadays, though antibiotics are available and scrub typhus outcomes have significantly improved, many patients die from this disease each year. The main reason is that diagnosing scrub typhus is challenging and misdiagnosis happens easily, leading

IP T

to delayed treatment and fatal organ failure.

SC R

The clinical signs, including acute undifferentiated febrile illness, headache,

myalgia, breathlessness, lymphadenopathy and varying involvement of organs such as the liver, lung and kidney (Janardhanan et al., 2014), are not pathognomonic and

U

usually misdiagnosed as other diseases, such as upper respiratory tract infectious

A

N

disease, herpes simplex virus, lymphad enitis, intestinal obstruction and pneumonia

M

(Du et al., 2016). Moreover, the disease usually happens in rural areas where the medical condition is limited and Weil–Felix testing with poor sensitivity and specificity

ED

is still used. The misdiagnosis rate is up to over 50% in China according to recent

PT

reports. In Shizong County, Yunnan Province, 22 cases of scrub typhus were reported

CC E

in 10 years, of which 15 cases were misdiagnosed (Man and Xu, 2017). The misdiagnosis rate was as high as 68.18%. Twenty-four cases with scrub typhus from January 2015 to December 2015 in Affiliated People's Hospital of Jiangsu University

A

were analyzed retrospectively and 18 cases were misdiagnosed as other diseases with a misdiagnosis rate of 75% (Du et al., 2016). Ninety-eight in 178 cases (about 55%) of scrub typhus were reported to be misdiagnosed from August 2010 to December 2015 in People’s Hospital of Suqian, Jiangsu Province of China (Wu et al., 2017), in which, 4

Weil-Felix testing was proved to recognize only 41.67% and 62.12% of the patient sera in 1 and 2 weeks post infection, respectively. The diagnosis of scrub typhus now relies on serological and molecular diagnostic tests. Weil-Felix test, indirect immunofluorescent assay (IFA), and ELISA are the

IP T

most commonly used serological diagnostic method. Weil–Felix test has been in use

SC R

for many years, though it is neither sensitive nor specific (Janardhanan et al., 2014).

IFA needs fluorescent microscopy and cultured O. tsutsugamushi antigen, and is not available in common laboratories. ELISA shows good sensitivity and specificity (Jang

U

et al., 2003, Prakash et al., 2006) while remains a time-intensive process. Moreover,

A

N

The serological testing method dependents on antibodies, and the significant IgM or

M

IgG antibody titer is usually detectable in 1 to 2 weeks post-infection (Xiong et al., 2012), which will delay appropriate treatment. The molecular diagnostic tests, such as

ED

polymerase chain reaction (PCR) or real-time quantitative PCR (RT-qPCR) are

PT

sensitive and specific, and allows the infection to be diagnosed as it is happening to

CC E

the patient. However, the instruments remain too expensive for routine diagnosis in an resources-limited endemic areas and the operation is too complex. Recombinase polymerase amplification (RPA) assay is a kind of isothermal DNA amplification

A

method and the reaction can be done in 5 to 20 min at 37 oC (Piepenburg et al., 2006). The amplification products with biotin and carboxyfluorescein (FAM)-modified could be visually observed with the naked eyes using a lateral flow (LF) test. The combination of RPA and LF (RPA-LF) lowers the instrument requirement for 5

molecular diagnostic tests and suits the diagnosing conditions in the field, in simple and crude removable laboratories, or in some basic medical units such as county or township hospitals. Considering the wide prevalence and high misdiagnosis rate of scrub typhus in

IP T

basic medical units of China, this study established a rapid, simple, and visual

SC R

molecular diagnostic test towards a Chinese epidemic strain of O. tsutsugamushi using RPA assay and LF test.

U

Materials and Methods

A

N

Culture and purification

M

The Chinese epidemic strain Sanjie of O. tsutsugamushi isolated previously (Cao et al., 2016) was cultured in L929 cells. After over 90% of the cells were infected, O.

ED

tsutsugamushi were purified using a modified Percoll gradient purification method

PT

(Ko et al., 2013). The purified bacteria were resuspended in Sucrose phosphate glutamate buffer (SPG, 218 mM sucrose, 3.8 mM KH2PO4, 7.1 mM K2HPO4, 4.9

CC E

mM monosodium L-glutamic acid) and stored at -80oC until use.

A

Samples and DNA

The animal samples used in this study have been mentioned in previous

published paper (Cao et al., 2016). Briefly, 60 small animals, including Apodemus agrarius, Rattus norvegicus, Microtus fortis, and Neomys fodiens were collected 6

annually from 2009 to 2013 in a sports school in Anhui Province of northern China. O. tsutsugamushi was detected in organs of 5 of the 60 animals by PCR in that study. DNAs from organs of these 5 infected animals and 55 uninfected animals were used in the present study. DNAs from organs of 2 Kunming mice experimentally infected

IP T

by O. tsutsugamushi (Cao et al., 2016) were also used. Genomic DNAs of Coxiella

SC R

burnetii, Rickettsia rickettsii, Rickettsia heilongjiangensis, Rickettsia sibirica were kindly given by Professor Bohai Wen. Bacteria of Staphylococcus aureus, and

Streptococcus suis were kindly given by Research Institute for Medicine of Nanjing

U

Command. Genomic DNAs of O. tsutsugamushi, S. aureus and S. suis were extracted

A

N

from the corresponding bacteria using QIAamp Blood and Tissue Mini DNA kit

M

(Qiagen, CA, USA). For quality control, genomic DNAs of R. rickettsii, C. burnetii, R. heilongjiangensis, R. sibirica, S. aureus, and S. suis were detected using RT-qPCR as

ED

described previously (Gong et al., 2015, Niu et al., 2008, Qi et al., 2013, Xiong et al.,

PT

2017, Xiong et al., 2014, Yu-Xin et al., 2010, Zhu et al., 2012) to be 105 copies/μL to

CC E

108 copies/μL.

For human samples, 20 mL of blood samples were collected from cubital veins of

healthy volunteers and centrifuged at 2000×g for 10 min to separate plasma. Various

A

of concentrations of genomic DNAs of O. tsutsugamushi were mixed with each of 200 μL of the plasma to make DNA-spiked samples. DNAs of spiked samples or normal plasma were extracted using a QIAamp Blood and Tissue Mini DNA kit (Qiagen). Concentrations of DNAs of O. tsutsugamushi in DNA-spiked samples were 7

quantified using RT-qPCR as described previously (Zhu et al., 2006). DNAs of normal plasma were detected using PCR targeting GAPDH gene (Tan et al., 2008) to make sure the DNA was extracted successfully.

IP T

Primers and probe

The 56-kDa outer membrane protein gene (Genebank: KM095135) was selected

SC R

as the target gene. PCR primers were designed with restriction enzyme cutting sites of BamH I or EcoR I (Table 1). The primers and probe for RPA were designed manually

U

indicated in Table 1, including 3 forward primers, 4 reverse primers, and 1 probe. The 5’

N

end of the reverse primers was labeled with biotin. The 5’ end of the probe was labeled

M

A

with Carboxyfluorescein (FAM), the 3’ end blocked with a phosphate group, and a base analog tetrahydrofuran (THF) inserted between the 30th and 31st base. All primers and

PT

ED

probes were synthesized by Genscript Company of Nanjing (Jiangsu Province, China).

CC E

Construction of recombinant plasmid

The 56-kDa outer membrane protein gene was amplified with conventional PCR

A

using primers indicated above and genomic DNA of O. tsutsugamushi as template, using a Premix Taq Version 2.0 kit (Takara, Dalian, China). The amplified products were analyzed by agarose gel electrophoreses and purified using a QIAquick Gel Extraction kit (Qiagen). Plasmid pUC19 was purified from Escherichia coli cells 8

using a TaKaRa MiniBEST Plasmid Purification Kit. Both the purified amplified products and plasmid pUC19 were digested using restriction enzymes BamH I and EcoR I and linked using a DNA Ligation Kit of Takara as per the manufacturer’s instructions, following with transformation of competent E. coli cells. After selected

IP T

on a solid Luria-Bertani (LB) medium with ampicillin, recombinant plasmid

SC R

56kDa-pUC19 in the positive bacterial colonies was purified, digested using both

BamH I and EcoR I, and analyzed by agarose gel electrophoresis. The recombinant plasmid was also sequenced (Genscript Company of Nanjing). The concentration of

U

the recombinant plasmid was measured using Nanodrop 2000 (Thermo Fisher

A

N

Scientific, Shanghai, China), and the copies were calculated as follows:

M

concentrations (copies/μL) = 6.02×1023 × plasmid concentration (ng/μL) × 10-9 / (DNA

ED

length × 660).

PT

Screening of primers

Three forward primers combined with 4 reverse primers made 12 combinations

CC E

as shown on Table 2. A commercial TwistAmp® RPA nfo kit (TwistDx™ Limited, Cambridge, UK) was used to screen primers and establish the RPA assay. The

A

preliminary reaction system recommended by the manufacturer’s instruction was used to screen the best combination. Briefly, 2.1 μL of forward primer (10 μM), 2.1 μL of reverse primer (10 μM), 0.6 μL of probe (10 μM), 1 μL of template (recombinant plasmid 56kDa-pUC19 or control plasmid pUC19 at a concentration of 1×104 9

copies/μL), 12.2 μL of DNase- and RNase-free water, and 29.5 μL of rehydration buffer were mixed together and added to the reaction tubes to rehydrate the pellet containing freeze-dried recombinase, polymerase, and single-strand binding protein. Then 2.5 μL of MgAc (280 mM) was added to the tube cap. To initiate the reaction,

IP T

MgAc in the cap was spun down into the tube, vortexed, and incubated at 37 oC for 20

SC R

min. After 4 min of incubation, the tube was vortexed again to improve the reaction efficiency.

After the amplification, 5 μL of the products were mixed with Tris-buffered

U

saline to a total of 100 μL in a well of a 96-well plate. Millenia Genline Hybridetect-1

A

N

(MGH) strips (Millenia Biotec GmbH, Gieben, Germany) were used for analysis of

M

the amplified products. The sample pad of the strip was immersed into the dilution for 3 to 5 min and the result was determined visually according to the test line (T line)

ED

and control line (C line) on the strip. Briefly, as a positive result, both the T line and C

PT

line were developed, indicating the DNA labeled with both FAM and biotin existed in

CC E

the amplicons; only the C line being developed indicated a negative result; only the T line being developed indicated the strip should be replaced. A modest primer group was screened when the T line in the experimental

A

reaction (with 56kDa-pUC19 as templates) was developed intensively and that in the control reaction (with pUC19 as templates) wasn’t developed.

10

Optimization of RPA-LF assay

To optimize the RPA assay, various concentrations of reverse primer (10μM, 5μM, or 2.5μM) and probe (10μM, 5μM, or 2.5μM) were used to conduct the RPA-LF

IP T

method as mentioned above. Recombinant plasmid 56kDa-pUC19 and control plasmid pUC19 at a concentration of 1×104 copies/μL were used as experimental and

U

was visually judged from the lines developed on the strips.

SC R

control template, respectively. After the reaction, the best concentration combination

N

Moreover, various amplification time (10min，15min，and 20min) was evaluated

A

in the RPA assay to determine the best reaction time; various volumes of amplification

M

products (1 μL, 2 μL, or 5 μL) were used to develop the strips to determine the best

ED

loading volume. All the reactions above were conducted in duplicate.

PT

Evaluation of detection limit and specificity

CC E

Both recombinant plasmid 56kDa-pUC19 and genomic DNA of O. tsutsugamushi were used as templates to evaluate the detection limit of the optimized

A

RPA-LF method.

For recombinant plasmid, a series of concentrations from 1 to 1×104 copies/μL

were diluted, and the dilutions were used as templates to evaluate the detection limit. For genomic DNA, a series of concentrations were diluted and quantified using 11

RT-qPCR as described previously. (Zhu et al., 2006). Meanwhile, the dilutions were used as templates to evaluate the detection limit of the optimized RPA-LF method. For specificity analysis, genomic DNAs of O. tsutsugamushi, R. rickettsii, C. burnetii, R. heilongjiangensis, R. sibirica, S. aureus, and S. suis were used as

IP T

templates to evaluate the optimized RPA-LF method.

SC R

Each reaction above was conducted in duplicate.

Detection of animal samples and simulative clinical samples

U

DNAs purified from spleens of O. tsutsugamushi-infected or uninfected mice

N

were detected by both RT-qPCR and the established RPA-LF method. The results

M

A

were compared and analyzed.

DNAs from the simulative clinical samples of genomic DNA-spiked plasma

ED

were quantified their concentrations of genomic DNAs of O. tsutsugamushi using

PT

RT-qPCR. Ten simulative clinical samples with concentrations between 10 to 50 copies/μL (experimental group) as well as 10 samples of normal human plasma DNA

A

CC E

(control group) were used to evaluate the established RPA-LF method.

Results

Construction of recombinant plasmid

The partial gene of 56-kDa outer membrane protein was amplified with 12

conventional PCR and linked to pUC19 plasmid to construct a recombinant plasmid 56kDa-pUC19, which was digested by enzymes BamH I and EcoR I and analyzed by agarose gel electrophoresis. As shown on Fig 1, two fragments were developed on the bands at sizes of about 3,000 bp and 750 bp, which accorded with the sizes of plasmid

IP T

pUC19 and the target gene, respectively. The target gene in the recombinant plasmid

SC R

was also sequenced to be consistent with sequence of KM095135 in GeneBank.

Screening of primers

U

Three forward primers and 4 reverse primers were combined to make 12

N

combinations, which were screened using the preliminary RPA-LF reaction system.

M

A

As a result (Fig 2), group 10 using forward primer OtF299 and reverse primer OtR583 performed best with an intensive band on the T line of the experimental strip and no

ED

band on the T line of the control strip. This group of primers were screened for

PT

establishing the RPA assay.

CC E

Establishment and optimization of RPA-LF method

The RPA-LF method was established using the screened primers. The

A

concentrations of reverse primer and probe, amplification time, and loading volume of amplified products on MGH strips were optimized. Various concentrations of reverse primer and probe indicated on Table 3 were used to carry out the RPA assay and the amplified products were developed using the 13

MGH strips. As shown on Fig 3A, groups 1, 2, and 4 developed bands with color obviously more intensive than the other groups on the T line of experimental strips. Comparing the primer and probe concentrations of these 3 groups comprehensively, group 4, with a lower concentration of probe resulting in a lower cost, was selected as

IP T

the best group. The optimized concentrations of reverse primer and probe were 10 μM

SC R

and 5 μM, respectively.

RPA assay was carried out for 10, 15, or 20 min, and the amplified products were detected using the MGH strips (Fig 3B). Results showed that the amplified products

U

could be detected after amplification for 15 min while an amplification time of 20 min

A

N

led to a better result with more intensive color of band.

M

The loading volumes of amplified products for development of the strips were

ED

evaluated and 5 μL of amplified products led to a better sensitivity (Fig 3C).

PT

Detection limit and specificity

The detection limit of the optimized RPA-LF method on detection of

CC E

recombinant plasmid 56kDa-pUC19 and genomic DNA of O. tsutsugamushi was evaluated. As shown on Fig 4A, as low as 10 copies of plasmid 56kDa-pUC19 could

A

be detected by the RPA-LF method. A series of concentrations of genomic DNA were diluted and detected by RPA-LF (Fig 4B) and RT-qPCR (Fig 4C and 4D). Both the methods could detect genomic DNA dilutions No. 2 to No. 5 but not No. 1 (Fig 4B and 4D), indicating both methods shared similar detection limit of about 12 14

copies/reaction quantified in dilution No. 2. The color depth of the T line was proportional to the concentrations of the templates. For specificity evaluation, the optimized RPA-LF method could detect genomic

sibirica, S. aureus, or S. suis shown on Fig 5, indicating the method is O.

SC R

tsutsugamushi-specific.

IP T

DNA of O. tsutsugamushi but not R. rickettsii, C. burnetii, R. heilongjiangensis, R.

Evaluation of the RPA-LF method with animal samples and simulative

U

clinical samples

N

DNA samples from organs of 62 animals, including 5 O. tsutsugamushi-infected

M

A

animals trapped in the wild, 2 mice infected in the lab, and 55 uninfected animals trapped in the wild, were used to evaluate the established RPA-LF method. All the 7

ED

samples from O. tsutsugamushi-infected animals were visually detected by RPA-LF

PT

method and the 55 uninfected samples were not (Fig 6A). The 7 infected samples were quantified as 1601 copies/μL to 2373 copies/μL and the 55 uninfected samples

CC E

were not detectable by RT-qPCR. Ten simulative clinical samples with concentrations between 10 to 50 copies/μL

A

(experimental group) as well as 10 samples of normal human plasma DNA (control group) were detected using the established RPA-LF method. As shown on Fig 6B, all experimental samples were detectable by RPA-LF method. However, the T line of a control sample (strip 11 on Fig 6B) was developed with a very light color band. The 15

sensitivity and specificity of the method in detecting simulative clinical samples were 100% and 90%, respectively. These results indicate a good accordance exists in sensitivity and specificity

IP T

between the established RPA-LF method and RT-qPCR.

SC R

Discussion

In previous study, Chao et al. have established RPA-nfo (RPA-LF) and RPA-exo

U

detection methods targeting the 47-kDa gene towards various strains of O.

N

tsutsugamushi (Chao et al., 2015). The sequences of 47-kDa protein have been

A

compared to show greater than 96% identity between strains with the exception of one

M

strain, O. chuto (Izzard et al., 2010), and is suitable to be selected as a target gene to

ED

develop universal detection method towards most of O. tsutsugamushi strains. However, it’s reported the 47-RPA-exo and 47-RPA-nfo needed further improvement

PT

to achieve lower detection limits in order to be more clinically applicable since the

CC E

detection limit ranged from 10 to 400 copies/reaction in detection of various strains (Chao et al., 2015). Therefore, in this study, a type-specific 56-kDa outer membrane

A

protein gene was used towards a Chinese epidemic Karp-like strain. In fact, the most commonly used molecular biology-based method is a nested PCR targeting partial sequence of the 56-kDa gene (Janardhanan et al., 2014), and a few studies have reported that conventional PCR targeting the 56-kDa gene shows high sensitivity and 16

specificity (Izzard et al., 2010). The target sequence with a length of 285 bp between the screened forward and reverse primers (OtF299 and OtR583) shares 93.6% identity with that of Karp strain (Genebank: M33004) in this study, while 76.8%, 74.8%, and 76.5% identity with that of strain Taiwan (Genebank: DQ485289; a Gilliam-like strain,

IP T

for the 56-kDa gene of Gilliam was not found in Genebank), Kato (Genebank:

SC R

M63382), or Kawazaki (Genebank: M63383), evaluated by DNAMAN software

(Version 5.2.2) as shown on Fig 7. It’s believed that this partial sequence is more type-specific in distinguishing different strains considering the huge difference in the

U

probe sequence.

A

N

In optimization of the RPA-LF, concentrations of both reverse primer and probe

M

but not forward primer were evaluated. In the instructions of the TwistAmp® RPA nfo kit, we notice that, in the reaction of RPA, 2 rounds of amplification exist. Amplified

ED

products of the first round using forward and reverse primer were used as templates

PT

for the second round of amplification. The probe, which anneals with the amplified

CC E

product and is cut off its 3’ block by nfo enzyme, will function as forward primer in the latter amplification. The first round of amplification may play a pick-up and enrichment role and the second round of amplification produces products detectable

A

by lateral flow strips. Without the first round, the reaction system still works considering there are target DNAs in the samples to be detected. So it’s presumed the forward primer may play less important roles than the reverse primer or the probe. Also considering the reverse primer and probe with modifications cost more than the 17

forward primer, concentration of the forward primer was not optimized. Finally a lower concentration of probe was used in the RPA assay for economic purpose. The established RPA-LF method could detect as low as 10 copies/reaction of positive plasmid or 12 copies/reaction of genomic DNA. The whole reaction,

IP T

including amplification and strips development, consumes less than 30 min. The

SC R

method is very specific without recognizing genomic DNA from phylogenetically

related organisms, like R. rickettsia, R. sibirica, C. burnetii, etc. Evaluated by limited samples from infected animals with high concentration of DNA copies, or by

U

simulative clinical samples with low concentration (a little higher than the detection

A

N

limit), both the sensitivities were 100% and the specificities were 100% and 90%,

M

respectively. This is generally comparable to that of RT-qPCR considering they have a similar detection limit and without cross reaction with other organisms. However,

ED

RPA-LF is more attractive. A cheap heating block is the only needed device instead of

PT

an expensive fluorescent quantitative machine in RT-qPCR, which means a lot for

CC E

many laboratories or basic medical units with limited instruments and infrastructure. In some extreme conditions, even no device is needed, considering the reaction temperature is 37oC and human body temperature is an alternative constant

A

temperature supplier. This has been proved to be available by Chao et al. previously (Chao et al., 2015). More over, less time consuming and suitable for detection in the field makes it more promising. Judged from Fig 4A and 4B, the color depth of the T line was proportional to the 18

concentrations of the templates but not sensitive enough between 10 and 100 copies/μL or between 1×103 and 1×104 copies/μL of templates. So the quantity in the template could be roughly but not accurately determined. Maybe in the future, RPA-LF method could be developed to a quantitative method when the lateral flow

IP T

test is more sensitive.

SC R

In this research, we didn’t collect any clinical samples from patients of scrub

typhus, though animal or simulative human samples were used to evaluate the method. There will be many interference factors in actual clinical usage and the established

U

method still need more evaluation using clinical samples. Also the method is

A

N

type-specific, which is not suitable for detection other types of O. tsutsugamushi

M

except Karp and limits the application scope. More type- or region-specific RPA-LF methods towards O. tsutsugamushi are in need in the future.

ED

In conclusion, we successfully established an RPA-LF detection method towards

PT

a Chinese epidemic Karp-like strain. The method consumes less than 30 min and

CC E

results can be determined visually with a detection limit of 10 and 12 copies/reaction in detection of positive plasmid and genomic DNA respectively, and a modest specificity without recognizing other organisms. The method is promising for

A

wide-ranging use in the basic medical units of the epidemic area, though more clinical patient samples are needed to evaluate the method in the future.

19

Acknowledgments We also thank Professor Bohai Wen for his kindly and generously providing DNA of several kinds of organisms.

IP T

Author Contributions Yong Qi and Yuexi Li, study design and writing; Qiong Yin and Yinxiu Shao,

SC R

data collection; Min Cao , Suqin Li, Hongxia Chen, Wanpeng Shen, and Jixian Rao,

sample collection; Jiameng Li, Xiaoling Li, Yu Sun, Yu Lin, Yi Deng, Wenwen Zeng,

U

Shulong Zheng, and Suyun Liu, data analysis.

A

N

Declarations of interest

ED

Funding Source

M

None.

PT

This research was funded by National Natural Science Foundation of China (31600151 and 81472932), Natural Science Foundation of Jiangsu Province

CC E

(BK20160130), Social Development Foundation of Jiangsu Province (BE2016622), China Key Project of New Medicine Development (2014ZX09304314-003), and

A

Medical Science and Technology Projects (17QNP052 and 15MS164).

Ethical Approval All animal experiments and the use of human blood samples were approved by

20

the Ethics Committee of Huadong Research Institute for Medicine and Biotechniques. The methods were carried out in accordance with the approved guidelines. Consent

IP T

form was signed.

References Section A. Informa Healthcare 1994.

SC R

Beran GW, Steele JH. Handbook of Zoonoses: Bacterial, Rickettsial, Chlamydial and Mycotic

Cao M, Che L, Zhang J, Hu J, Srinivas S, Xu R, et al. Determination of Scrub typhus Suggests a New Epidemic Focus in the Anhui Province of China. Sci Rep 2016;6:20737.

Chao CC, Belinskaya T, Zhang Z, Ching WM. Development of Recombinase Polymerase

U

Amplification Assays for Detection of Orientia tsutsugamushi or Rickettsia typhi. Plos Negl Trop Dis 2015;9(7):e0003884.

N

Du S, Ding K, Zhang Y, Sun C. Analysis of misdiagnosis of 18 cases with tsutsugamushi disease.

A

Chin J Lepr Skin Dis 2016;32(12):736-7.

Gong W, Qi Y, Xiong X, Jiao J, Duan C, Wen B. Rickettsia rickettsii outer membrane protein YbgF

M

induces protective immunity in C3H/HeN mice. Human Vaccines 2015;11(3):642-9. Izzard L, Fuller A, Blacksell SD, Paris DH, Richards AL, Aukkanit N, et al. Isolation of a novel 2010;48(12):4404.

ED

Orientia species (O. chuto sp. nov.) from a patient infected in Dubai. Journal of Clinical Microbiology Janardhanan J, Trowbridge P, Varghese G. Diagnosis of scrub typhus. Expert Rev Anti Infect Ther 2014;12(12):1533-40.

PT

Jang WJ, Huh MS, Park KH, Choi MS, Kim IS. Evaluation of an Immunoglobulin M Capture Enzyme-Linked Immunosorbent Assay for Diagnosis of Orientia tsutsugamushi Infection. Clinical & Diagnostic Laboratory Immunology 2003;10(3):394.

CC E

Ko Y, Choi J, Ha N, Kim I, Cho N, Choi M. Active escape of Orientia tsutsugamushi from cellular

autophagy. Infect Immun 2013;81(2):552-9. Man M, Xu Z. Analysis of the causes of clinical misdiagnosis of tsutsugamushi disease and its

A

epidemiology in 15 cases. Chinese Community Doctors 2017. Niu D, Yang X, Chen M. Rapid detection of spotted fever group rickettsiae with real-time

quantitative PCR. Medical Journal of Chinese Peoples Liberation Army 2008. Piepenburg O, Williams C, Stemple D, Armes N. DNA detection using recombination proteins.

PLoS biology 2006;4(7):e204. Prakash JA, Abraham OC, Mathai E. Evaluation of tests for serological diagnosis of scrub typhus. Tropical Doctor 2006;36(4):212-3. Qi Y, Xiong X, Duan C, Jiao J, Gong W, Wen B. Recombinant protein YbgF induces protective 21

immunity against Rickettsia heilongjiangensis infection in C3H/HeN mice. Vaccine 2013;31(48):5643. Rosenberg R. Drug-resistant scrub typhus: Paradigm and paradox. Parasitology Today 1997;13(4):131. Tan XW, Jing-Hua XU, Wen HB. Construction of Recombinant Plasmid of Human GAPDH Gene. Journal of Nanhua University 2008. Wu C, Yao C, Wang G, Sun M. Analysis of Diagnosis and Treatment for Misdiagnosed Patients with Tsutsugamushi Disease. Clinical Misdiagnosis & Mistherapy 2017;30(3):51-4. Xiong X, Jiao J, Gregory AE, Wang P, Bi Y, Wang X, et al. Identification of Coxiella burnetii CD8+ Mouse Model. Journal of Infectious Diseases 2017;215(10):1580.

IP T

T-Cell Epitopes and Delivery by Attenuated Listeria monocytogenes as a Vaccine Vector in a C57BL/6 Xiong X, Qi Y, Jiao J, Gong W, Duan C, Wen B. Exploratory study on Th1 epitope-induced protective immunity against Coxiella burnetii infection. Plos One 2014;9(1):e87206.

SC R

Xiong X, Wang X, Wen B, Graves S, Stenos J. Potential serodiagnostic markers for Q fever

identified in Coxiella burnetii by immunoproteomic and protein microarray approaches. BMC Microbiology,12,1(2012-03-16) 2012;12(1):35.

Xu G, Walker DH, Jupiter D, Melby PC, Arcari CM. A review of the global epidemiology of scrub

U

typhus. Plos Neglected Tropical Diseases 2017;11(11):e0006062.

Yu-Xin SU, Gao S, Kang L, Zhao Y, Zheng XL, Wang JL. Establishment of real-time quantitative

N

PCR-based methods for detection of Staphylococcus aureus in food. Bulletin of the Academy of Military Medical Sciences 2010;34(1):25-4.

A

Zhu J, Zhang JH, Dan HU, Shi J, Zhang XY, Xian-Fu LI, et al. Establishment of multiplex real-time

M

PCR to detect highly virulent Streptococcus suis serotype 2. Journal of Pathogen Biology 2012. Zhu L-n, Zhang J-b, Chen M-l, Wen B-h, Niu D-s, Li Q-f, et al. Detection of Orientia tsutsugamushi

A

CC E

PT

ED

by quantitative real-time PCR assay. CHINESE JOURNAL OF ZOONOSES 2006;22(3):228-31.

22

U

SC R

IP T

Figure Legends

N

Fig 1 Agarose gel electrophoresis analysis of double enzymes digested

A

recombinant plasmid 56kDa-pUC19. M, DNA marker; P, digested recombinant

A

CC E

PT

ED

M

plasmid 56kDa-pUC19.

23

IP T SC R U N

A

Fig 2 Screening of the best primers from 12 groups of primer combinations.

M

Primer combinations from 1 to 12 were indicated in Table 2. Recombinant plasmid

ED

56kDa-pUC19 and plasmid pUC19 were used as experimental template and control template, respectively. Group 13 is a control using template, primers, and probe given

A

CC E

PT

in the TwistAmp® RPA nfo kit (TwistDX™ Limited, Cambridge, UK).

24

IP T SC R U N A M ED PT CC E

Fig 3 Optimization of the RPA-LF method. A, nine combinations of reverse

primer and probe in various concentrations indicated on Table 3 were used to carry

A

out the RPA assay and the amplified products were developed using the MGH strips; B, RPA assay was carried out for 10, 15, or 20 min, and the amplified products were developed using the MGH strips; C, various volumes of amplified products (1, 2, or 5 μL) were used to develop the MGH strips. Recombinant plasmid 56kDa-pUC19 (P) 25

and plasmid pUC19 (N) at a concentration of 1×104 copies/μL were used as

ED

M

A

N

U

SC R

IP T

experimental and control templates, respectively.

PT

Fig 4 Detection limit analysis of the RPA-LF method. A, various concentrations

CC E

of plasmid 56kDa-pUC19 were used to evaluate the detection limit of RPA-LF method; B, various concentrations of genomic DNA of O. tsutsugamushi were used to

A

evaluate the detection limit of RPA-LF method; C, standard curve of the RT-qPCR used to determine the concentrations of genomic DNA of O. tsutsugamushi; D, the amplification plot of RT-qPCR which was used to determine the concentrations of genomic DNA of No. 1 to 5 in B to be undetected, 12 copies/μL, 91 copies/μL, 1013 26

A

N

U

SC R

IP T

copies/μL, and 12179 copies/μL, respectively.

M

Fig 5 Specificity analysis of the RPA-LF method. Templates used in 1 to 7 were

ED

O. tsutsugamushi, Rickettsia rickettsii, Coxiella burnetii, Rickettsia heilongjiangensis,

A

CC E

PT

Rickettsia sibirica, Staphylococcus aureus, and Streptococcus suis, respectively.

27

IP T SC R U N A

M

Fig 6 Representative results of RPA-LF in detecting O. tsutsugamushi-infected

ED

animal samples (A) or simulative patient plasma samples (B). A, strips 1 to 13 show results of detecting uninfected mouse samples and strips 14 to 20 show results of

PT

detecting infected mouse samples; B, strips 1 to 10 show results of detecting samples

CC E

of genomic DNA of O. tsutsugamushi–spiked human plasma and strips 11 to 20 show

A

results of detecting samples of normal human plasma.

28

IP T SC R

Fig 7 Multiple alignment of the selected sequence with that of various strains including Karp, Taiwan, Kato, and Kawazaki using DNAMAN software. The identity

A

CC E

PT

ED

M

A

N

U

was indicated.

29

Table 1 Primers and probe for PCR and RPA Nucleotide sequence (5’-3’)

Primers or probe FP

AAGGATCCAGAGCAGAGCTAGGGTTTATG

RP

GCGAATTCTATACCCATCAAAAAAATCTCTG

OtF299

GGATCGCTTGGTTAAAGAATTGTGCTGGTA

OtF321

TGCTGGTATTGACTATAGGGTAAAAGATCC

OtF379

AATCCGGTGTTGTTAAATATTCCACAGGGT

OtR493

Biotin-TATACCTCCATTGCTCATGGTTATGTATCG

RPA Reverse

OtR520

Biotin- TTGATAATGCAGCAAGTCCAATTACCATAT

primers

OtR546

Biotin- AGGAGGATCGATAGGTTTATTAGCATTTGA

OtR583

Biotin- CCTTCCTCTGAGTAATTTCATCAGTTAATA

PCR primers

RPA Forward

SC R

IP T

primers

FAM-ATTCCACAGGGTAACCCTAATCCTGTTGGA-[TH RPA Probe

OtProbe

M

A

N

U

F]-ATCCACCGCAGCGAG-PO4

Group No.

Primer pairs

Group No.

Primer pairs

OtF299+ OtR493

7

OtF299+ OtR546

OtF321+ OtR493

8

OtF321+ OtR546

OtF379+ OtR493

9

OtF379+ OtR546

4

OtF299+ OtR520

10

OtF299+ OtR583

5

OtF321+ OtR520

11

OtF321+ OtR583

6

OtF379+ OtR520

12

OtF379+ OtR583

2

A

CC E

3

PT

1

ED

Table 2. Primer groups combined with various primers.

30

Table 3 Groups of OtR583 and OtProbe using various concentrations Concentrations of

Group

Concentrations of

Group

Concentrations of

No.

OtR583&OtProbe

No.

OtR583 and OtProbe

No.

OtR583 and OtProbe

1

10μM&10μM

4

10μM&5μM

7

10μM&2.5μM

2

5μM&10μM

5

5μM&5μM

8

5μM&2.5μM

3

2.5μM&5μM

6

2.5μM&5μM

9

2.5μM&2.5μM

A

CC E

PT

ED

M

A

N

U

SC R

IP T

Group

31