Rv3737 is required for Mycobacterium tuberculosis growth in vitro and in vivo and correlates with bacterial load and disease severity in human tuberculosis

Background Rv3737 is the sole homologue of multifunctional transporter ThrE in Mycobacterium tuberculosis (Mtb). In this study, we aimed to investigate whether this transporter participates in vitro and in vivo survival of Mtb. Methods To characterize the role of Rv3737, we constructed and characterized a Mtb H37RvΔRv3737. This strain was evaluated for altered growth rate and macrophage survival using a cell model of infection. In addition, the comparative analysis was conducted to determine the association between Rv3737 mRNA expression and disease severity in active pulmonary TB patients. Results The H37RvΔRv3737 strain exhibited significantly slow growth rate compared to H37Rv-WT strain in standard culture medium. Additionally, the survival rate of H37Rv-WT strain in macrophages was 2 folds higher than that of H37RvΔRv3737 at 72 h. A significantly higher level of TNF-α and IL-6 mRNA expression was observed in macrophages infected with H37RvΔRv3737 as compared to H37Rv-WT. Of note, Rv3737 expression was significantly increased in clinical Mtb isolates than H37Rv-WT. The relative expression level of Rv3737 was positively correlated with lung cavity number of TB patients. Similarly, the higher Rv3737 mRNA level resulted in lower C(t) value by Xpert MTB/RIF assay, demonstrating that a positive correlation between Rv3737 expression and bacterial load in TB patients. Conclusions Our data takes the lead in demonstrate that the threonine transporter Rv3737 is required for in vitro growth and survival of bacteria inside macrophages. In addition, the expression level of Rv3737 may be associated with bacterial load and disease severity in pulmonary tuberculosis patients. Supplementary Information The online version contains supplementary material available at 10.1186/s12879-021-06967-y.

of reactivation of latent TB infection [4]. Specially, individuals with immunosuppression have higher odds of TB reactivation compared with normal individuals [5]. Therefore, Mtb has evolved many strategies to survive long periods of stress encountered in the human host by reducing its metabolic activity and modulation of the host immune response.
Bacteria are equipped with a broad variety of transport systems [6]. Transport processes play a pivotal role in pathogen metabolism, e.g. for the uptake of nutrients and excretion of harmful agents [7]. Moreover, several amino acid transporters are reported as being associated with pathogenesis [6]. For instance, threonine transporter ThrE is essential for the fitness of Corynebacterium glutamicum in vitro growth [8]. Inactivation of thrE gene shows reduced growth rate in vitro in medium supplemented with threonine. In addition to threonine, the ThrE carrier serves to export small molecules, indicating that it is a multifunctional transporter that gets rid of metabolic waste products and thus hold more importance to thrE in biological fitness [8].
Of note, only two homologues are identified in Mtb and S. coelicolor [8]. Rv3737 encodes a 55 Da protein with significant sequence similarity to characterized ThrE protein [9]. As a member of a new translocator family that has never been reported before, it is interesting to investigate whether this multifunctional potential transporter participates in vitro and in vivo survival of Mtb. To characterize the role of Rv3737, we constructed and characterized an Mtb Rv377 deletion mutant (H37RvΔRv3737). This strain was evaluated for altered growth rate and macrophage survival using a cell model of infection. In addition, the comparative analysis was conducted to determine the association between Rv3737 mRNA expression and disease severity in active pulmonary TB patients.
In addition, 12 clinical Mtb isolates were collected from a set of sputum smear-positive and GeneXpert MTB-positive specimens from August 2016 to February 2017. The demographic and clinical characteristics were available from electronic medical records. The number of cavities in the lungs is obtained by reading the images of the patient's chest CT examination. This study was subject to approval by the Ethics Committees of the Affiliated Hospital of Zunyi Medical University. All patients were≥18 years old and provided written informed consent prior to enrolment.

Construction of H37RvΔRv3737
Using genomic DNA of H37Rv-WT as a template, the Rv3737 nucleic acid sequence was derived from Genbank of NCBI (https:// www. ncbi. nlm. nih. gov/ gene/ 885794). As shown in Fig. 1A, the primers for the upper and lower arms of the Rv3737 gene and verification primers were designed based on the principle of homologous recombination (Additional file 3: Table S2). Flanking regions comprising upstream and downstream regions of the Rv3737 gene were amplified by PCR and cloned into the p0004S plasmid containing a hygromycin resistance cassette, the vector was then ligated into the phAE159 plasmid, which was electroporated into M. smegmatis mc 2 155, and the resulting phage was amplified to obtain a high-titer stock. The high-titer phage was utilized to infect H37Rv-WT, which was plated onto selective 7H10 agar plates. Plates were incubated for 4-8 weeks at 37 °C, which eventually gave rise to the growth of a small number of H37RvΔRv3737 colonies. Colonies were picked, and PCR and qPCR were used to confirm the presence of the hygromycin-gene flanking region and the absence of the Rv3737 gene (Fig. 1B, C) [10,11].

Generation of Msm/pMV261-Rv3737
The full-length sequence of Rv3737 was amplified from H37Rv-WT genomic DNA. Primers for transcriptional level analysis are listed in Additional file 3: Table S2. The fragment was inserted into the downstream of promoter hsp60 within the recombinant pMV261 plasmid. The constructed plasmid was introduced into M. smegmatis mc 2 155 by electroporation, and the presence of the vector was confirmed by Western Blot. Empty plasmid pMV261 was electroporated into M. smegmatis mc 2 155, but without Flag tag (Fig. 1D).

Growth and colonial morphology of Rv3737 knockout strain
H37RvΔRv3737 was inoculated in selective 7H9 broth medium and incubated at 37 °C. Optical density at 600 nm (OD 600 ) was detected at intervals of 24 h and the growth curves at 37 °C were obtained. H37RvΔRv3737 was inoculated in selective 7H10 agar plates. After incubation for 3 weeks at 37 ℃, the colony morphology was recorded with the HP scanner. The H37Rv-WT was used as a control and the same treatment was performed. For scanning electron microscope analysis, the bacteria were collected by centrifugation at 500 rpm for 5 min. Then 2.5% glutaraldehyde was added into the pellet for 24 h for fixation purpose. Followed by treatment with 1% osmium tetroxide and ethanol gradient, the samples were sprayed with gold film and observed in a SU8010 scanning electron microscope (Hitachi, Japan) [12].

High performance liquid chromatography (HPLC)
The concentration of threonine in the growth medium was quantified with High Performance Liquid Chromatography. Briefly, the freshly cultured mycobacteria, (16 days for Mtb and 4 days for Msm), were centrifuged at 12,000g for 15 min, and the supernatant were obtained and filtered with a disposable 0.22 μm cellulose acetate. After derivatization, the mixture was transferred into a 100 µL glass insert in an amber vial and analyzed by HPLC (Rigol L3000-system, Rigol, Beijing, China) as described by Priscila del Campo et al. [13]. Norleucine was used as an internal standard for assessing the recovery of amino acid from liquid medium.

Survival of Rv3737 knockout strain and Rv3737-overexpressing strain in RAW 264.7
H37RvΔRv3737 and H37Rv-WT were infected to 5 × 10 5 / well RAW 264.7 cells in 6-well plate at multiplicity of infection (MOI) 10. After 4 h of incubation, all extracellular bacteria were removed gently by washing and intracellular bacteria were harvested. At 24 and 72 h after infection, both extracellular bacteria released from macrophage lysed in supernatant and intracellular bacteria in intact cell layer were harvested. Bacteria at 4 h, 24 and 72 h were plated on 7H10 agar plates in triplicate, plates were incubated for 3 weeks at 37 °C and CFU (colony forming unit) were counted [14].
Msm/pMV261-Rv3737 and Msm/pMV261 were infected to 5 × 10 5 /well RAW 264.7 cells in 6-well plate at multiplicity of infection (MOI) 10. After 4 h of incubation, all extracellular bacteria were removed gently by washing and intracellular bacteria were harvested. At 0 h, 6 h, 12 h, 24 and 48 h after infection, both extracellular bacteria released from macrophage lysed in supernatant and intracellular bacteria in intact cell layer were harvested. Bacteria at 0 h, 6 h, 12 h, 24 and 48 h were plated on 7H10 agar plates in triplicate, plates were incubated for one week at 37 °C and CFU (colony forming unit) were counted.

Cytokine measurement
Culture supernatants and sediments from Mtb-infected RAW 264.7 cells were harvested at 0 h, 4 h, 8 h, 12 h, 24 h post-infection and stored at − 80 °C for cytokine measurement. The concentrations of TNF-α and IL-6 in culture supernatant were detected using an enzyme-linked immunosorbent assay (ELISA) kit according to the manufacturer's instructions (Solarbio, Beijing, China) [15]. The mRNA level of TNF-α and IL-6 in culture sediments were determined using qPCR. In simple terms, the total RNA was extracted with Trizol method according to the instructions of the manufacturers [16]. After treatment with DNaseI (TaKaRa, Dalian, China), the cDNAs were reverse-transcribed from 5 µg of total RNA with the Pri-meScript ™ II 1st Strand cDNA Synthesis Kit (TaKaRa, Dalian, China). Real time PCR (qPCR) was carried out in triplicates for each sample using TB Green ® Premix Ex Taq ™ II (TaKaRa, Dalian, China) in the CFX96 touch Real-Time PCR System (Bio-Rad) [17]. Primers for transcriptional level analysis are listed in Additional file 3: Table S2. GAPDH were used as the internal control of the respective qPCR experiments.

Rv3737 expression of clinical isolates and H37Rv-WT
The Mtb isolates at log phase were lysed by ultrasound and then subjected to RNA extraction. The mRNA level of Rv3737 in clinical isolates and H37Rv-WT was determined using qPCR as method above. Primers for transcriptional level analysis are listed in Additional file 3: Table S2 and SigA were used as the internal control of the respective qPCR experiments.

Sputum collection, bacterial growth and bacterial load measurement in sputum
Sputum specimens were collected for acid-fast staining and GeneXpert MTB/RIF assay (Cepheid, Sunnyvale, CA, USA). Acid-fast staining microscopy was performed directly on all samples as described previously [18]. One milliliter sputum was mixed with 2 ml sample reagent, and incubated at room temperature for 15 min. Then the decontaminated sample was then added to a test cartridge and loaded onto the Xpert instrument. Results were reported as the cycle threshold (Ct) values that represented the minimal PCR cycles required for detection threshold [19]. The average Ct values of five probes were used to estimate bacterial load after exclusion of any delayed values due to rifampicin resistance [20]. 1.0 ml of sputum specimen with positive results from both tests were treated with N-acetyl-L-cysteine-NaOH-Na citrate (2.00% final concentration). After neutralization and centrifugation, the suspension of the pellet was inoculated on Lowenstein-Jensen (L-J) medium. The visible growth of colonies on L-J medium was identified using conventional biochemical method [14].

Statistical and analysis
GraphPad Prism v7.03 (GraphPad Software, San Diego, CA, USA) was used to analyze the data and generate graphs. t test was utilized to compare two groups of data, two-way ANOVA was used for three or more groups of data, considering p value < 0.05 to be significant. The linear relationships were analysed by the R squared correlation method. Spearman Coefficient was conducted to establish the relationship between the expression level of Rv3737 and the Ct value yielded by Xpert and between the expression level of Rv3737 and the number of cavities.

.
At the amino acid level, Rv3737 shared 29.60% sequence identity with ThrE of C. glutamicum ( Fig. 2A). We have constructed H37RvΔRv3737 and Msm/ pMV261-Rv3737 at the same time. The concentration of threonine in culture medium were compared across different strains (Fig. 2B, C). Our data revealed that the free threonine concentration of the culture medium where the H37RvΔRv3737 (0.85 ± 0.06 µg/ml) grew was remarkedly decreased compared to that of H37Rv-WT strain (1.23 ± 0.08 µg/ml, p < 0.05), while the overexpression of Rv3737 resulted in a significant elevation of threonine concentration in M. smegmatis (1.30 ± 0.10 µg/ml for Msm/pMV261-Rv3737 verses 1.01 ± 0.03 µg/ml for Msm/ pMV261, p < 0.05). We then assessed whether inactivation of Rv3737 affects the in vitro growth and physiology of Mtb. As shown in Fig. 3A, the H37RvΔRv3737 strain exhibited significantly slow growth rate compared to H37Rv-WT strain in standard culture medium. Colony size of H37RvΔRv3737 and H37Rv-WT strain was compared by plating the same dilution on plates after 21 days (Fig. 3B, C). The average colony size of H37RvΔRv3737 was 0.38 ± 0.02 μm, which was much smaller than that of H37Rv-WT strain (0.58 ± 0.02 μm, Fig. 3D).
Field emission scanning electron microscopy was conducted to assess cell morphology of Mtb, and measured the length and width of the cell with Image J, each strain has no less than 120 bacteria. Individual cells from two strains were indistinguishable in morphology characteristics, whereas significant differences were observed in cell lengths. As noted in Fig. 4, the average cell length of the H37Rv-WT and H37RvΔRv3737 were 1.89 ± 0.04 and 1.69 ± 0.05 μm (p < 0.05), respectively, suggesting that the latter had a shorter cell length. In contrast, the average cell width of H37RvΔRv3737 was 0.41 ± 0.01 μm, which was statistically higher than that of H37Rv-WT strain (0.35 ± 0.01 μm, p < 0.05). was used to align sequences and reconstruct a phylogenetic tree based on sequence similarities. Shared amino acids are highlighted in red selected for testing in this study. B The free threonine concentration of H37Rv-WT and H37RvΔRv3737 in 7H9 medium grown for 16 days. C The free threonine concentration of Msm/pMV261 and Msm/pMV261-Rv3737 in 7H9 medium grown for 4 days. The difference between the wild type and the mutant was significant by Student's t test (*p < 0.05; **p < 0.01; ***p < 0.001)

Rv3737 promotes the survival of Mtb in macrophages
Mouse macrophages were infected with H37Rv-WT, H37RvΔRv3737, Msm/pMV261, and Msm/pMV261-Rv3737 to determine differences in capacity for intracellular growth. As shown in Fig. 5A, B, the intracellular survival was assessed at 24 and 72 h after infection at 4 h as reference. The survival rate of H37Rv-WT strain was 2 folds higher than that of H37RvΔRv3737 at 72 h, respectively. However, the survival of Msm/pMV261-Rv3737 in macrophages was significantly increased relative to that of Msm/pMV261. Taken together, these data indicated that Rv3737 had an important role in enhancing the infection capability and intracellular survival of tubercle bacilli.

Rv3737 inhibits the release of TNF-α and IL-6 in macrophages
To explore the potential role of Rv3737 in modulating the innate immune response, we investigated the levels of cytokines upon infection of RAW264.7 cells with H37Rv-WT and H37RvΔRv3737. A significantly higher level of TNF-α and IL-6 mRNA expression was observed in macrophages infected with H37RvΔRv3737 as compared to H37Rv-WT (Fig. 5C, D). Detection of cytokines in the culture supernatants of macrophages also supported the elevated secretion of proinflammatory cytokines (TNF-α and IL-6) at increasing time points (Fig. 5E, F).

Correlation between Rv3737 expression and disease severity
Based on the slower in vivo growth and increased host proinflammatory cytokine response, we hypothesized that the expression level of Rv3737 correlated with Mtb virulence in the host. In order to test this hypothesis, we recruited 12 clinical Mtb isolates to determine whether the upregulation of Rv3737 could lead to more severe clinical symptoms. A total of 12 patients infected with diagnosed TB were retrospectively included in our analysis ( Table 1). As shown in Fig. 6A, Rv3737 expression was significantly increased in clinical Mtb isolates than H37Rv-WT. Of note, the relative expression level of Rv3737 was positively correlated with lung cavity number of TB patients (r = 0.71, p < 0.01, Fig. 6B). Similarly, the higher Rv3737 mRNA level resulted in lower C(t) value by Xpert MTB/RIF assay, demonstrating that a positive correlation between Rv3737 expression and bacterial load in TB patients (r = − 0.81, p < 0.01, Fig. 6C).

Discussion
Transporter systems are commonly considered as a potential tool for delivery of therapeutic agents. Recently experimental studies reveal that several transporters be required for chronic infection and expression of virulence in pathogenic bacteria [21,22]. In this study, we attempted to understand the possible role of transporter protein Rv3737, which alters in vitro growth and intracellular survival of bacteria inside macrophages. Depletion of Rv3737 in Mtb resulted in decreased growth rate in vitro compared to H37Rv-WT; however, the mutant strain displayed usual rough and dry colonies as control strain, indicating that Rv3737 might be not involved in the cell wall lipid remodeling in Mtb. Although the exact characteristics and role of this transporter in growth profile remain unclear, we speculate that the inactivation of Fig. 3 In vitro growth of H37RvΔRv3737. A Growth curve of H37Rv-WT and H37RvΔRv3737 in 7H9 medium. B Colonies of H37Rv-WT and H37RvΔRv3737 on 7H10 agar grown for 21 days. C Diameter of colonies of H37Rv-WT and H37RvΔRv3737 on 7H10 agar. Experiments were performed in triplicates. The difference between the wild type and the mutant was significant by Student's t test (*p < 0.05; **p < 0.01; ***p < 0.001) Fig. 4 Morphological changes between the H37Rv-WT and H37RvΔRv3737 strains. A-D Scanning electron microscope photographs of strains. Bacteria were grown in 7H9 medium. E Length of H37Rv-WT and H37RvΔRv3737. F Width of H37Rv-WT and H37RvΔRv3737. The difference between the wild type and the mutant was significant by Student's t test (*p < 0.05; **p < 0.01; ***p < 0.001) Rv3737 might lead to accumulation of metabolic waste products in vivo and consequently inhibit their growth.
Experimental evidence from previous studies confirms that highly virulent Mtb isolates have faster in vivo doubling time [23,24]. We found that the knockout of Rv3737 had a markedly negative impact on intracellular survival as compared to the control. On one hand, this fact may reflect the declined growth rate of tubercle bacilli in macrophages, as demonstrated in vitro observations. On the other hand, the elevated levels of proinflammatory cytokines were noted in the culture supernatant of macrophages infected with H37RvΔRv3737, thus promoting intracellular bacteria clearance in macrophages.
Specific mechanism behind this significant correlation is presently unclear. On the basis of its putative transporter function, the molecules exported by Rv3737 into extracellular substance were able to impair host defense against intracellular bacteria via inhibiting inflammatory response. On the basis of our findings, Rv3737 may participate in modulation of reduced or delayed host proinflammatory cytokine response, which is required for persisting virulence and survival of Mtb within host macrophages. Furthermore, the elevated expression level of Rv3737 was noted in clinical Mtb isolates as compared to H37Rv-WT with attenuated virulence. This diversity supports E, F Immunoassays for TNF-α and IL-6 in supernatants collected from macrophages infected with H37Rv-WT and H37RvΔRv3737. The difference between the wild type and the mutant was significant by Student's t test (*p < 0.05; **p < 0.01; ***p < 0.001) our previous findings that Rv3737 may be involved in virulence of Mtb. Notably, a significant positive correlation between Rv3737 expression level and bacterial load in pulmonary TB patients raises the possibility that the isolates with increased expression of Rv3737 are prone to escape clearance by alveolar macrophages, thus leading to greater bacillary multiplication in the host. The high bacterial burden always causes more lung damage and higher mortality [25,26]. In line with previous observation, we observed that the higher expression of Rv3737 was most likely to result in more cavities among patients affected by pulmonary TB. In view of the strong association between Rv3737 and lung pathology, we speculate that is could be used as a candidate biomarker for predicting the virulence of distinct isolates, and its potential pathogenic effect in host. We also acknowledged several obvious limitations to the present study. First, we observed the slower growth rate and shorter length of H37RvΔRv3737; whereas the mutant had larger width than H37Rv-WT control, which may be associated with the remodeling of cytoskeleton to modulate the bacterial growth. However, the reason for this phenomenon remains unclear. Second, despite validating the potential role of Rv3737 as transporter of threonine in Mtb, we were unable to successfully construct the Mtb strains with overexpression of Rv3737, indicating the expression balance of this gene was subtly regulated in tubercle bacilli. As an alternative, the Rv3737-overexpressed M. smegmatis model was used to testify its function for mycobacteria. Third, in vitro findings were not verified with animal model, which weakened the significance of conclusion. Fourth, the pathological pattern of pulmonary TB is a complicated process between host and pathogen, and multiple factors may bias our conclusion, such as genetic lineage of Mtb, immunology status of host and sample size of validation cohort. Thus the primary correlation between the expression of Rv3737 and pathological patterns is required to confirm in the future. Finally, although we recruited active TB cases in our analysis, the diverse courses of TB disease across patients may serve as a major challenge that biases our conclusion.

Conclusions
In conclusion, our data firstly demonstrate that the threonine transporter Rv3737 is required for in vitro growth and survival of bacteria inside macrophages. In addition, the expression level of Rv3737 may be associated with bacterial load and disease severity in pulmonary tuberculosis patients. Further studies will be conducted to elucidate the molecular mechanism of this transporter to Mtb virulence.