Detection of Pneumocystis jirovecii and Toxoplasma gondii in patients with lung infections by a duplex qPCR assay.

Pneumocystis pneumonia (PCP) and pulmonary toxoplasmosis (PT) are caused by Pneumocystis jirovecii and Toxoplasma gondii. The clinical symptoms and imaging of PCP and PT are indistinguishable. A duplex qPCR was developed to differentiate between these two pathogens. In testing 92 clinical samples to validate the performance of this method for P. jirovecii detection, it identified 31 positive samples for P. jirovecii infection, consistent with clinical diagnosis. Among the remainder of the 61 clinical samples with suspected PCP, yet showing as negative by the conventional PCR diagnosis approach, 6 of them proved positive using our new assay. Our new approach also produced similar results in identification of T. gondii infections, giving a result of 2 positive and 20 negative in clinical samples. An investigation was undertaken on the prevalence of P. jirovecii and T. gondii infections using 113 samples from lung infection patients. 9% (10/113) were shown to be positive with infections of P. jirovecii, 2% with T. gondii (2/113) and 5% (6/113) were co-infected with both pathogens. Although this duplex qPCR can detect individual P. jirovecii and T. gondii infection, and co-infection of both pathogens, further large-scale investigations are needed to validate its performance, especially in T. gondii detection. Our assay provides a rapid and accurate tool for PCP and PT diagnosis in immunocompromised population and clinical surveillance of these infections in patients with no immune defects.

Introduction

Pneumocystis jirovecii (previously called P. carinii) and Toxoplasma gondii are opportunistic pathogens that can cause pneumocystis pneumonia (PCP) and pulmonary toxoplasmosis (PT), respectively, in immunocompromised patients [1,2]. With the success of antiviral therapy in HIV patients, PCP and PT are now being more commonly observed in patients with other immune deficiency rather than AIDS [3,4], such as patients that had received organ transplantation [5-7], the elderly [8] and infants with suppressed or immature immunity [9,10]. P. jirovecii infection may be involved in colonization previously or transmission by person to person [11]. T. gondii can infect through ingestion of cysts from raw or badly-cooked meat, oocysts from substrates contaminated with the feces of infected felines and congenital transmission by tachyzoites [12]. However, these infections are easily neglected or misdiagnosed as other diseases due to the non-specific clinical symptoms and manifestations. In addition, the imaging characteristics of PCP and PT are very similar to each other, they are not distinguishable by clinical lung imaging only. Misdiagnosis of PCP or PT would result in the delayed and insufficient treatment, leading to rapid disease progression with high mortality. The mortality rates for PCP are 4% in AIDS patients and 27% for non-AIDS related patients [4]. While for PT, the mortality rate can reach 43.5% for patients with previous allogeneic haematopoietic stem cell transplantation treatment [3]. Immunocompromised patients are often co-infected with both P. jirovecii and T. gondii. The lung infections of these two pathogens are usually manifested mainly as an uncontrolled pneumonia with life threatening outcomes [13-16]. As early diagnosis can improve the clinical outcome, a rapid, accurate and robust tool for P. jirovecii and T. gondii diagnosis is urgently needed in clinical practice.

In addition to the analysis of clinical manifestations, imaging, epidemiology and immunocompromised conditions, the accurate and definitive diagnosis of PCP and PT depends on the identification of pathogens using immunological, molecular-biological or microscopic methods [17,18]. Unfortunately, the existing serological antibody tests show cross-reactions with other pathogens and cannot distinguish between current and historical infections. The 1-3-β- glucan is a serum biomarker for PCP diagnosis, however, it is not specific and the cut-off value for PCP diagnosis is not known, limiting its use in clinical practice [19]. Direct fluorescent antibody assay is a reliable method for P. jirovecii identification while less sensitivity than molecular methods [20]. The microscopic examination for the presence of cysts of P. jirovecii and T. gondii in the sputum or bronchoalveolar lavage fluid (BALF) under microscopy is considered as the gold standard for definitive diagnosis of P. jirovecii and T. gondii infections. However, this needs an experienced technician or parasitologist as the P. jirovecii trophozoites are easily missed under microscope, especially when the infection burden is low [21,22]. Parasite culturing of T. gondii can increase the sensitivity of identification. However, it is time-consuming to culture T. gondii in vitro. Furthermore, P. jirovecii cannot be successfully cultured in vitro. In addition, conventional PCR, which use mitochondrial large subunit rRNA (mtLSU) as target gene, is currently used in clinical practice [23]. However, compared to conventional PCR, real-time quantitative PCR (qPCR) is more convenient for clinical applications and its results are more reliable.

qPCR plays an important role in P. jirovecii and T. gondii detection, which can not only specifically identify the presence of pathogen DNA but also provide a quantitative measurement of the pathogen load [24]. Previous studies have reported qPCR detection of P. jirovecii based on the mtLSU, NADH dehydrogenase subunit 1 (NAD1), Cytochrome b (CYTB), and Dihydropteroate synthase (DHPS) genes [25]. Among these targets, mtLSU gene was the first one for PCP molecular diagnosis and is still the most widely used [26-30]. For T. gondii detection, qPCR assays have also been reported with various targets, such as B1, Rep529, COX1, ITS1 [31-33]. However, these previously reported assays are single target qPCR assays designed for the detection of only one pathogen. In this study we developed a duplex qPCR method, targeting the multicopy mitochondrial small subunit rRNA (mtSSU) gene (37 copies per genome) of P. jirovecii and the repetitive 529 fragment (Rep-529) (200–300 copies per genome) of T. gondii for the simultaneous detection of both pathogens in patients with lung infections.

Materials and methods

Ethics statement

This project has been approved by the Ethics Committee of Beijing Friendship Hospital (Beijing, China) with approval number of 2020-P2-193-01. All clinical samples investigated in this study were obtained from an existing sample collection. All samples were anonymized.

Patients and clinical and DNA samples

A total of 227 clinical samples from patients at Beijing Friendship Hospital, Capital Medical University were enrolled in this study from Sep 2019 to Sep 2020 (Table 1). The sputum, BALF and blood specimens were collected from them for the duplex qPCR testing. Clinical diagnostic criteria: PCP diagnosis is based on the analysis of the clinical manifestation, imaging characteristics and conventional PCR which amplifies P. jirovecii mtLSU gene from sputum and BALF specimens [23]. PT diagnosis is based on the analysis of clinical manifestation, imaging characteristics and positive serological IgM and/or IgG [18].

Table 1

Characteristics of samples.

Category	Groups	Age	Gender (M%)	Sample type				Conventional PCR		T. gondii immunoassay		clinical diagnostic criteria or results	Sputum or BALF culture results
				DNA	Sputum (n)	BALF (n)	Blood (n)	Positive (n)	Negative (n)	IgM (n)	IgG (n)
Clinical samples for Validation (n = 114)	PCP (n = 31)	62 (4 months, 84 years)	68%	-	14*	17	0	31	0	-	-	PCP diagnosis is based on the analysis of the clinical manifestation, imaging characteristics and conventional PCR which amplifies P. jirovecii mtLSU gene from sputum and BALF specimens [23].	-
	Suspected PCP (n = 61)	63 (4 months, 98years)	66%	-	48*	13	0	0	61	-	-
	T. gondii infections (n = 2)	6 and 56 years	100%	-	0	0	2	-	-	1	1	PT diagnosis is based on the analysis of clinical manifestation, imaging characteristics and positive serological IgM and/or IgG [18]. Due to lack of commercial kits of qPCR, pathogenic diagnosis relies on immunological methods.
	Non-T. gondii infections (n = 20)	34 (1, 73) year	70%	-	0	0	20	-	-	0	0
Clinical samples for Survey (n = 113)	Infant (n = 20)	4 (0, 10) days	85%	-	20#	0	0			-	-	Neonatal pneumonia; Neonatal sepsis; Tetralogy of fallot; Neonatal wet lung; Respiratory infection	Alpha hemdytic streptococcus; Neisseria; Enterobacter cloacae; Klebsiella oxytoca
	Elder (n = 76)	79 (55, 104) years	61%	-	58*	8	0			-	-	Pneumonia; Lung cancer; Respiratory infections	Alpha hemdytic streptococcus; Neisseria; Enterobacter cloacae; Monilia; Acinetobacter baumannii; Klebsiella pneumoniae; Candida tropicalis; Corynebacterium; Stenotrophomonas maltophilia; Pseudomonas aeruginosa
	Other (n = 17)	48 (3,53) years	59%	-	14*	3	0			-	-	Pneumonia; Respiratory infections	Alpha hemdytic streptococcus; Neisseria, EB virus; Candida albicans; Candida smoothing; Monilia; Stenotrophomonas maltophilia
DNA samples (n = 8)	Positive control	-	-	P. jirovecii mtSSU plasmid; T. gondii rep-529 plasmid; T. gondii parasites DNA	-	-	-	-	-	-	-	-	-
	Negative control	-	-	alpha hemolytic Streptococcus; Neisseria spp; Mycoplasma;Mycobacterium tuberculosis; Klebsiella pneumoniae	-	-	-	-	-	-	-

All data are presented as median (ranges), BALF: bronchoalveolar lavage fluid,—present no data

* represent standard sputum

# represent induced sputum

The 114 samples were used to validate the performance of qPCR assay (Table 1). For P. jirovecii detection, 31 specimens (14 sputum and 17 BALF) from clinically diagnosed PCP patients confirmed by conventional PCR and 61 samples (48 sputum and 13 BALF) from clinically suspected PCP patients but negative in conventional PCR. For T. gondii detection, 2 blood samples from patients clinically confirmed with T. gondii infection, and 20 control blood samples collected from patients with defined malaria (2), dengue fever (1), brucellosis (1) and leishmaniasis (1), or non-defined infection, but all negative when tested with T. gondii ELISA kit.

Further 113 samples (103 sputum and 10 BALF) from patients with lung infections were used to investigate the prevalence rate of P. jirovecii and T. gondii infections (Table 1), including 20 infants, 76 elders and 17 others.

Eight DNA samples were applied in this study (Table 1). P. jirovecii mtSSU/pUC19 plasmid, T. gondii rep-529/pUC19 plasmid and DNA samples of T. gondii parasites are used as positive controls. Negative controls include DNA from Mycobacterium tuberculosis, alpha hemolytic Streptococcus, Neisseria, Mycoplasma and Klebsiella pneumoniae. In addition, one standard sputum was used for analytical sensitivity analysis.

Primers and probes design and construction of standard plasmids

Fourteen mtSSU gene sequences from different P. jirovecii strains and 37 Rep-529 gene sequences from different T. gondii strains from GenBank were aligned individually using the “Bioedit” software (v7.0.1, Ibis Biosciences, Carlsbad, CA, USA) (S1 and S2 Tables). Primers and probes were designed using Primer Express 3.0 and Oligo 7.0 software based on the conserved region of mtSSU of P. jirovecii and Rep-529 region of T. gondii with expected amplification size124bp for P. jirovecii and 233bp for T. gondii (Table 2). The specificity of the primers and probes was verified using NCBI Primer blast. The primers and probes were synthesized by Sangon Biotech Co., Ltd.

Table 2

Sequence of primers and probes designed for the duplex real-time quantitative PCR.

Species	Target gene	The target gene name	Primer and probe	Sequences	Amplicon size	GeneBank accession No.
Pneumocystis jirovecii	mtSSU	Mitochondrial small subunit ribosomal RNA	PJ-F	5’-TTATGAAGTGGGCTACAGAC-3’	124bp	JX499143
			PJ-R	5’-CTTCAAAGAGCCGAGTTCC-3’
			PJ-probe	5’-FAM-TCCGACTTCCATCATTGCATC-TAMRA-3’
Toxoplasma gondii	Rep-529	529-bp repeat element	TG-F	5’-GACTACAGACGCGATGCC-3’	233bp	DQ779189
			TG-R	5’-CTCTTCAATTCTCTCCGCCAT-3’
			TG-probe	5’-Texas Red-ACACCGGAATGCGATCTAGACGA-BHQ2-3’

mtSSU: Mitochondrial small subnuit ribosomal RNA; Rep-529: repetitive 529bp fragment

The mtSSU and Rep-529 target sequence fragments were PCR amplified using the primers described above from the clinical specimens from patients with PCP and cultured T. gondii parasite respectively, and purified with a DNA purification kit (TIANGEN, DP214, Beijing, CHN), These were then subcloned into plasmid pUC19 (TAKARA, 3219, Tokyo, Japan) using EcoRI and HindIII sites. The correct inserts of target DNA in the recombinant plasmids were confirmed by DNA sequencing.

DNA extraction

DNA was extracted from fresh sputum or BALF specimens collected from hospitalized patients. The specimens were pre-treated using 1N NaOH (1:1(v/v) for sputum, and 2:1(v/v) for BALF) 60°C for 1h to ensure complete liquefying before centrifugation at 8000 rpm for 5 min. The pellet was washed twice with saline and DNA was extracted from the pellet using a kit (TIANGEN, DP705, Beijing, CHN) according to the manufacturer’s instruction and stored at -80°C for further duplex qPCR assay.

DNA was extracted from blood samples using a DNeasy Blood & Tissue Kit (Qiagen, 69506, Hilden, GER) according to manufacturer’s instructions, which were used as controls for the validation of duplex qPCR assay.

Duplex quantitative real-time PCR assay

Duplex qPCR was conducted in a 20μl reaction containing 300nm each primers (PJ-F and PJ-R for P. jirovecii; TG-F and TG-R for T. gondii) and 200nm each probes (PJ-probe (5′ FAM/3′ TAMRA) for P. jirovecii; TG-Probe (5′ Texas Red/3′ BHQ2) for T. gondii with 10μl GoTaq Probe qPCR Master Mix (Promega, A6101, Madison, WI, USA) plus 1μl template DNA (5–50 ng). The reaction was performed in the Applied Biosystems 7500 Fast Real-Time PCR System (ABI) with 95°C for 2 min followed by 40 cycles of 95°C for 15 sec, 58°C for 50 sec. Each sample was tested with replicates, the plasmid mtSSU/pUC19 and Rep-529/pUC19 were used as positive control and reaction without template DNA (distilled water) was used as negative control in all experiments.

Analytical sensitivity and specificity

The limit of detection (LOD) was determined by minimum number of copies in samples (inter CV%<5%). Briefly, the clinical sample (standard sputum), which was P. jirovecii and T. gondii negative, was used as diluents, plasmids mtSSU/pUC19, Rep-529/ pUC19 and mtSSU/pUC19&Rep-529/pUC19 were added in liquefied standard sputum sample and diluted into 1000, 100, 10, 5, 3 or 1 plasmid copies/μl, respectively,the standard sputum sample as blank control. Total DNA were extracted. Detection assay was performed with three batches. The specificity of the duplex qPCR was compared with other DNA samples of obtained Mycobacterium tuberculosis, alpha hemolytic Streptococcus, Neisseria spp, Mycoplasma and Klebsiella pneumoniae.

Clinical sensitivity and specificity

Total 114 clinical samples (92 for P. jirovecii and 22 for T. gondii) were used to validate the performance of this new assay. DNA extraction and the duplex qPCR was performed according to the procedure described above and the detection results were compared with the clinical diagnosis.

Prevalence of P. jirovecii and T. gondii infections in lung infection patients

The duplex qPCR was further used to test 113 sputum and BALF specimens from lung infection patients (infants n = 20, elders n = 76 and others n = 17) to preliminary investigate the prevalence rate of P. jirovecii and T. gondii infections.

Statistical analysis

Statistical analysis was performed using the SPSS software version 20.0 and visualized on GraphPad Prism version 5.0. Continuous variables were presented as medians (ranges). The Mann-Whitney U-test was used to statistically compare between groups. P < 0.05 was considered as statistical significance.

Results

Specific detection of P. jirovecii mtSSU and T. gondii Rep-529 in a duplex PCR assay

Bioinformatics analysis was firstly performed to locate the potential targets which are conserved among different P. jirovecii (mtSSU, 99.78% identities) and T. gondii strains (Rep-529, 92.02% identity) (S1 and S2 Tables). Primers and probes were designed based on these conserved regions (Table 2). Further PCR tests with different DNA samples indicate these primers are highly specific to detect P. jirovecii and T. gondii and exhibit no cross-reactions with other pathogens, including Mycobacterium tuberculosis, alpha hemolytic Streptococcus, Neisseria, Mycoplasma and Klebsiella pneumoniae (Fig 1). Thus, this duplex PCR assay may be suitable to further establish qPCR system to detect both P. jirovecii and T. gondii simultaneously.

Fig 1

Specificity of the primers designed for P. jirovecii mtSSU and T. gondii Rep-529 amplification.

The duplex PCR method showed no cross-reaction to Mycobacterium tuberculosis, alpha hemolytic Streptococcus, Neisseria, Mycoplasma or Klebsiella pneumoniae.

Sensitivity of established duplex qPCR assay

A duplex qPCR assay was then established with the primers and probes described above, the LOD of which was 5 copies target DNA for P. jirovecii and 10 copies for T. gondii (S3 Table). The standard curves generated by serial copy number dilutions of plasmid DNA using single target qPCR and duplex qPCR, showed that the PCR efficiency in both PCR reactions was the same. The linear range for both were 8-log fold with a correlation coefficient (R2) of 0.998 (Fig 2). The intra- assay and inter assay coefficient of variation (CV) of quantification cycle (Cq) values for the 20 replicates was <3%, indicating high precision in the assay (S4 Table). The results indicated that the duplex qPCR assay has high sensitivity and precision to detect target P. jirovecii mtSSU and T. gondii Rep-529 genes, ideal for high throughput testing of clinical samples.

Fig 2

Quantitative correlation between gene copy number and threshold cycle of the duplex qPCR assay.

(A) P. jirovecii mtSSU plasmid was serially diluted from 101 to 108 copies/reaction and subjected to qPCR. (B) Linear regression of Cq vs. lg copy number of mtSSU plasmid using singleplex qPCR and duplex qPCR. (C) T. gondii Rep-529 plasmid was serially diluted from 101 to 108 copies/reaction and subjected to qPCR. (D) Linear regression of Cq vs. lg copy number of Rep-529 plasmid using singleplex qPCR and duplex qPCR. ΔRn = Rn (normalized reporter)-baseline. Ct, Cycle threshold.

Evaluation of duplex qPCR in the detection of P. jirovecii and T. gondii infections in clinical samples

Total 114 clinical samples were tested to validate the performance of this duplex qPCR assay (Tables 3 and S5). Firstly, the qPCR results revealed that 24 of 31 samples (14 sputum and 17 BALF) from patients with clinically diagnosed PCP were positive for P. jirovecii only, and 6 were detected as positive for P. jirovecii only in 61 suspected PCP samples. Secondly, it is interesting to note that 7 of 31 clinically diagnosed PCP samples are co-infected with T. gondii. Thirdly, 3 of 61 clinically suspected PCP samples were only positive for T. gondii. Two blood samples with antibodies (IgG/IgM) positive for T. gondii were positive for T. gondii DNA even though the detected copy number of Rep-529 gene was low (21 and 190/μl). In addition, twenty control blood samples were all negative for T. gondii, and all 22 blood samples were negative for P. jirovecii when the duplex qPCR was performed. The quantitative data showed that there was no significant difference in the amount of DNA between samples from sputum and BALF for P. jirovecii with DNA copies of 1.1×106 (7×103 to 2.8×109) (Fig 3A). T. gondii exhibited similar results with DNA copies of 7.03×104 (2.38×103 to 1.64×107) (Fig 3B).

Table 3

The duplex qPCR results for 227 clinical samples.

Category	Group	Duplex qPCR
		P. jirovecii positive only		T. gondii positive only		P. jirovecii & T. gondii positive			Negative number
		Number (%)	Cq (ranges)	Number (%)	Cq (range)	Number (%)	P. jirovecii Cq (ranges)	T. gondii Cq (ranges)
Clinical samples for Validation (n = 114)	PCP (n = 31)	24/31 (77%)	28.09 (19.24, 36.69)	0	-	7/31 (23%)	30.23 (25.38, 33.19)	32.43 (27.93, 35.61)	0
	Suspected PCP (n = 61)	6/61 (9.83%)	33.31 (27.26, 35.19)	3/61 (5%)	32.61 (24.99, 37.19)	0	-	-	52
	T. gondii infections (n = 2)	0	-	2	33.99 and 37.23	0	-	-	0
	Non-T. gondii infections (n = 20)	0	-	0	-	0	-	-	20
	Total (n)	30	-	5	-	7	-	-	72
Clinical samples for Survey (n = 113)	Infant (n = 20)	3/20 (15%)	34.55 (33.69, 35.94)	1/20 (5%)	38.28	3/20 (15%)	32.98 (31.03, 35.76)	32.01 (28.06, 33.52)	13
	Elder (n = 76)	7/76 (9%)	33.18 (32.76, 36.87)	1/76 (1%)	29.53	3/76 (4%)	30.94 (28.94, 31.17)	30.21 (29.72, 30.83)	65
	Other (n = 17)	0	-	0	-	0	-	-	17
	Total (n)	10	-	2	-	6	-	-	95

All data are presented as median (ranges), Cq: quantification cycle, n: number

Fig 3

The detected P. jirovecii (A) and T. gondii (B) DNA copy number comparison between samples from sputum and BALF.

The duplex qPCR was used to analyze another 113 samples from hospitalized patients with other lung infections (Tables 3 and S5). The results showed that 10 patients were positive for P. jirovecii only (8.8%), 2 patients were positive for T. gondii only (1.8%), and 6 patients were positive for both infections (5.3%) (Fig 4A). All infected patients were either the infants with age younger than 4 days or the elder with age over 55 years old. These infected patients include 7 infants with age of 1–4 days old (3 P. jirovecii, 3 P. jirovecii co-infection with T. gondii and 1 T. gondii) with infection rates of 30% for P. jirovecii infection (6/20) and 20% for T. gondii infection (4/20) that are higher than the infection rate in elder patients with age from 55 to 91 years old (7/76 for P. jirovecii, 1/76 for T. gondii and 3/76 for co-infection) (Fig 4B). Quantitative results showed that the copy number of P. jirovecii mtSSU in positive samples (n = 16) was 9.65×104 (6.1×103 to 2.1×106) copies/μl, and the copy number of T. gondii Rep-529 in positive samples (n = 8) was 3.05×105 (1.1×103 to 1.8×106) copies/μl (Fig 4C).

Fig 4

Detection of P. jirovecii and T. gondii infection using duplex qPCR assay on clinical samples from 113 lung infection patients.

(A) Number of P. jirovecii, P. jirovecii + T. gondii and T. gondii patients detected by the duplex qPCR. (B) P. jirovecii and T. gondii positive patients detected with duplex qPCR in infant and elder groups. (C) DNA load of P. jirovecii and T. gondii detected in lung infection patients.

Discussion

In our study, we designed a new duplex qPCR based on the conserved regions of the mitochondrial small ribosomal subunit (mtSSU) of P. jirovecii and the Rep-529 of T. gondii. The mtSSU rRNA of P. jirovecii is a multi-copy gene located in the mitochondria of the fungus, with as high as 37 copies in each fungus [25]. This allows for potential higher detection sensitivity compared to other target genes such as mtLSU rRNA (15 copies), NAD1 (23 copies), CYTB (6 copies), and DHPS (single copy). The high copy number of mtSSU rRNA targets in P. jirovecii, could possibly increase the chance to detect the pathogens even if there is a low infection burden. This target is also highly specific for P. jirovecii without cross-reaction with other pathogens (Fig 1). For T. gondii, the Rep-529 loci has been reported to be repeated 200- to 300-fold in the genome of T. gondii and yielded far higher sensitivity than amplification of B1 sequence (35copies), ITS (110 copies), 18SrDNA (110 copies), SAG (single copy), GRA1 (single copy) genes [25,33-35]. Evaluation of this duplex qPCR assay showed that this method could detect five copies of mtSSU of P. jirovecii and ten copies of the Rep-529 of T. gondii from standard sputum samples, and without cross-reaction with other common pathogens causing pneumonia, such as Mycobacterium tuberculosis, alpha hemolytic Streptocoocus, Neisseria, Mycoplasma, and Klebsiella pneumonia, Plasmodium, Dengue virus, Brucella and Leishmania spp. The high sensitivity and specificity of this duplex qPCR assay suggests that it can be used to diagnose PCP caused by P. jirovecii infection and PT caused by T. gondii infection.

To verify this, 92 clinical samples from patients with diagnosed/suspected PCP were used to validate the performance of this assay for P. jirovecii detection. The results showed that this assay could detect the fungal DNA in all 31 clinically diagnosed PCP samples confirmed by conventional PCR. In 61 samples from patients with suspected PCP based on the PCP clinical manifestation but negative with conventional PCR, 6 of 61 were positive when performed with duplex qPCR, indicating the duplex qPCR has higher sensitivity compared to the standard PCR assay against the mtLSU gene [36,37]. As mtSSU is 2.5 more copies vs mtLSU, it may be the key that this new qPCR assay exhibits higher sensitivity than the current standard method. As for the T. gondii detection, only two blood samples with clinically confirmed T. gondii infection (IgM/IgG positive) were used in this study. Considering limited samples size and no sputum or BALF samples were applied in this study, a larger sample size needs to be examined in future to add further confidence for the detection of T. gondii infections.

This new duplex qPCR method was used to determine the prevalence of P. jirovecii and T. gondii infections in 113 lung infection patients. The results showed that 8.8% of them (10/113) were infected with P. jirovecii, 1.8% with T. gondii (2/113) and 5.3% (6/113) were co-infected with both P. jirovecii and T. gondii. Firstly, this study demonstrates that this designed duplex qPCR can be used to detect not only for individual P. jirovecii and T. gondii infections, but also for co-infections in a single assay, with higher sensitivity than the single conventional PCR tests previously reported for individual qPCR assays for these pathogens [38-41]. Secondly, it suggested that infections of P. jirovecii and T. gondii are common in patients with general bacterial or virus lung infections, which showed no typical symptoms of pneumonia caused by these two pathogens. It is possibly because they have normal immunity that prevents the pathology caused by these opportunistic pathogens. However, we could not exclude if the lung infection is related to the pathogen infections since there was a lack of detail clinical information. Several studies reported that, colonization rate of P. jirovecii and T. gondii reached 37–55% and 19% in patients with lung diseases, respectively [42-44]. Indeed, our results indicated that the copy number of P. jirovecii DNA from clinical confirmed/suspected samples was significantly higher than that from survey group samples without typical PCP symptoms (P = 0.002) (Figs 3A and 4C). It will be important to determine the correlation between the pathology and the colonized pathogen load, and with the immunological status of the infected people, which needs to examine more specimens in further investigations. Thirdly, some neonates, initially diagnosed as Alpha hemdytic streptococcus, Neisseria or Escherichia coli infections, were also detected positive for P. jirovecii and/or T. gondii infections when this duplex qPCR was performed (Table 3). Other studies also reported that, the PCR positive rate of P. jirovecii and T. gondii were 15–32% and 9.9%, individually in rhinitis aspirate from hospitalized children (1 month-2 years) with bronchopneumonia or acute respiratory infection and umbilical cord tissue from neonates [45-47]. It indicated that P. jirovecii and/or T. gondii infections in neonates were neglected in clinical practice and it should be taken more serious attention in therapeutic management, especially in cases of prematurity, malnutrition and congenital infection. If the congenital toxoplasmosis is suspected or people with infections of P. jirovecii and T. gondii is confirmed, the usage of immunosuppressant such as corticoid should be avoided to prevent the diffused infections of these opportunistic pathogens in lungs.

Therefore, our assay provides a rapid and accurate tool for not only diagnosing PCP and PT patients in immunocompromised population, but also clinical surveillance of infections of P. jirovecii and T. gondii in patients without immune defects. While the sample size used in this study was relatively small (n = 113), we argue the results are very supportive that this assay is worth pursuing further for the prevalence of P. jirovecii and T. gondii infections. A larger cohort of patients should be tested in future to add further confidence in the clinical value of this valuable new diagnostic tool.

In general, P. jirovecii and T. gondii involvement in lung infection patients has been seriously underestimated, leading to misdiagnosis and often fatal consequences. Early diagnosis of these infections would reduce medical costs, morbidity and mortality. Our currently designed duplex qPCR detection system in this study exhibits highly sensitivity and specificity to detect P. jirovecii and T. gondii infections simultaneously for PCP and PT patients or for populations with no immune defects. It is a fast and simple assay to perform. Further large-scale investigations with more clinical samples should be performed to further evaluate its potential application in clinical practice.

mtSSU Alignment.

(XLSX)

Click here for additional data file.

Rep-529 Alignment.

(XLSX)

Click here for additional data file.

The sensitivity of duplex qPCR.

(XLSX)

Click here for additional data file.

Precision of intra and inter-assay of mtSSU and Rep-529-based qPCR assay.

(XLSX)

Click here for additional data file.

Duplex qPCR assay results.

(XLSX)

Click here for additional data file.