Leptospirosis is among the world’s most important zoonotic infectious diseases, characterized by variable manifestations ranging from asymptomatic or self-resolving acute febrile illness to severe disease with a combination of fever, acute kidney injury, jaundice, severe pulmonary hemorrhage syndrome, refractory shock, and aseptic meningitis. Important advances have been made in diverse aspects of leptospirosis including the differential host responses to leptospiral infection [4–9]. Despite these advances, mechanistic details by which end organ damage develops in some individuals but not in others remain to be elucidated. Further, factors that govern host susceptibility to leptospiral infection are not well understood. A recent study by our collaborative group found that approximately 6% of randomly sampled individuals in a highly endemic rural Amazonian village were chronically colonized by Leptospira without any recent clinical evidence of infection. Although renal colonization by leptospires may occur in humans without serological or clinical evidence of infection, the clinical relevance and functional consequences of leptospiral colonization in humans remain to be characterized. New, less invasive, less expensive and more practical tools (compared to kidney biopsy) are needed to study pathologic changes in the kidney in order to understand clinically relevant sequelae of infection. In this report, we use proteomic analysis of urine exosomes in a rat chronic colonization model as a non-invasive window to kidney function in asymptomatic leptospiral infection.
Exosomes are nanovesicles that are released from cells as a mechanism of intercellular communication. Characterization of exosomes from different biological samples has shown the presence of common as well as cell-type specific proteins. The protein content of exosomes has been shown to be modified under pathological or stress conditions [12–14]. Since exosome contents are specifically derived from cellular components, here we tested the hypothesis that urinary exosome protein content from a rat infected with Leptospira would be different from that of the uninfected rat, and that these differences would hold key information about the pathways mediating host responses to Leptospira infection.
After renal colonization, persistent shedding of Leptospira is clearly established in carrier animal hosts, especially rodents. However, they rarely develop symptoms and are not noticeably impaired by infection of their kidneys. We have recently detected chronic asymptomatic renal colonization by Leptospira in human subjects from a rural Amazonian village. We reasoned that the structural and functional changes in the kidney that arise following asymptomatic Leptospira infection are different from symptomatic disease. These differences between the asymptomatic and symptomatic leptospirotic kidneys can be understood by studying the downstream products of the kidney, such as urine.
Given the nephron cell-state-specific cargo of the urinary exosome, we hypothesized that urine exosome analysis holds key information that is relevant to differences between clinically symptomatic and asymptomatic leptospirosis infection. This report is the first step towards testing this hypothesis. Here we report our preliminary findings from the exosome proteomic analyses of urines from rats infected with Leptospira using uninfected rat urine exosome as controls. We also studied the host-response to Leptospira infection in male and female rats separately.
This work was approved by the Institutional Animal Care and Use Committee of the University of California, San Diego.
Experimental infection of animals and study design
Three-week old Sprague Dawley rats (Charles River Laboratories, USA) (6 male rats and 6 female rats) were housed in cages of 1–2 rats each, with food and water provided ad libidum. Six animals (3 males and 3 females) were inoculated intraperitoneally with 108 mid-log phase L. interrogans serovar Copenhageni strain HAI1026. Uninfected controls were inoculated ip with sterile EMJH. To determine the health status, weight, general body condition, posture, activity, appetite food and water consumption of each animal was monitored daily. Infection was confirmed by serology (microscopic agglutination test) and quantitative polymerase chain reaction (qPCR) of weekly urines, and Warthin-Starry silver staining of kidney sections following necropsy. In order to keep the minimum number of animals for the purpose of statistical calculations, we used this small number of animals. This is also discussed as a limitation of the study under discussion.
Urine sampling and processing
Urine from each animal was collected weekly starting 7 days after post-challenge by placing the animals individually for 20–24 hrs in metabolic cages, and was captured into containers containing Roche Complete Protease Inhibitor, one tablet per 5 mL urine. Urines were separately centrifuged at 3000 x g for 30 min. The supernatant was withdrawn, the pH adjusted to 7, aliquoted and frozen at -70°C until further analysis.
Genomic DNA was extracted from these weekly urines and qPCR was used to assess leptospiruria. In addition, at necropsy, kidneys were harvested, fixed in formalin, paraffin embedded, and infection confirmed by Warthin-Starry silver stain (S1 and S2 Figs.).
Exosome preparation and protein resolution analysis
Exosomes from terminal urine samples from infected and uninfected rats were prepared using an in-house protocol developed based on the solvent exclusion principle using polyethylene glycol (PEG)-induced precipitation. To prevent naturally occurring peptides in the exosome from confounding post in-gel trypsinization peptide information of the full-length proteins, we conducted 1 dimensional SDS-PAGE of the exosome proteins prior to in-gel trypsinization.
Each rat urine sample (only terminal sample) was separately analyzed, without pooling any sample. Each rat urine sample was run in a separate gel lane. Gel slices for each lane were cut to 1 mm x 1 mm cubes and destained 3 times by first washing with 100 μL of 100 mM ammonium bicarbonate for 15 min, followed by addition of the same volume of acetonitrile (ACN) for 15 min. The supernatant was transferred to a clean tube and samples lyophilized and reduced by mixing with 200 μL of 100 mM ammonium bicarbonate-10 mM DTT then incubated at 56°C for 30 min. The liquid was removed and 200 μL of 100 mM ammonium bicarbonate-55mM iodoacetamide was added to gel pieces, which were then incubated at room temperature in the dark for 20 min. After removal of the supernatant and one wash with 100 mM ammonium bicarbonate for 15 min, an equal volume of ACN was added to dehydrate the gel pieces. The solution was then removed and samples lyophilized.
For digestion, ice-cold trypsin (0.01 μg/μL) in 50 mM ammonium bicarbonate solution was added in enough amounts to cover the gel pieces and set on ice for 30 min. After complete rehydration, the excess trypsin solution was removed, replaced with fresh 50 mM ammonium bicarbonate, and left overnight at 37°C. The peptides were extracted twice by the addition of 50 μl of 0.2% formic acid and 5% ACN and vortexed at room temperature for 30 min. The supernatant was removed and saved. A total of 50 μL of 50% ACN-0.2% formic acid was added to the sample, which was vortexed again at room temperature for 30 min. The supernatant was removed and combined with the supernatant from the first extraction. The combined extractions from all the gel slices in a lane pertaining to a single rat urine exosome sample was separately analyzed directly by liquid chromatography (LC) in combination with tandem mass spectroscopy (MS/MS) using electrospray ionization. Thus, multiple slices of gels representing a single rat urine exosome sample was used for mass spectrometry. Two replicates per rat exosome sample were run on the MS.
Trypsin-digested mixtures were analyzed by the Eksigent nanoLC-Ultra 2D System (Eksigent, AB SCIEX Dublin, CA, USA) combined with cHiPLC-nanoflex system (Eksigent) in trap-elute mode. Briefly, samples were first loaded on the cHiPLC trap (200 μm x 500 μm ChromXP C18-CL, 3 μm, 120 Å) and washed in isocratic mode with 0.1% aqueous formic acid for 10 min at a flow rate of 3 μL/min. The automatic switching of cHiPLC ten-port valve then eluted the trapped mixture on a nano cHiPLC column (75 μm x 15 cm ChromXP C18-CL, 3 μm, 120 Å), through a 45 min gradient of 5–50% of eluent B (eluent A, 0.1% formic acid in water; eluent B, 0.1% formic acid in acetonitrile), at a flow rate of 300 nL/min. To preserve system stability, in terms of elution times of components, trap and column were maintained at 35°C.
Mass spectra were acquired using a QExactive mass spectrometer (Thermo Fisher Scientific, San José, CA, USA), equipped with a nanospray ionization source (Thermo Fisher). Nanospray was achieved using a coated fused silica emitter (New Objective, Woburn, MA, USA) (360 μm o.d./50 μm i.d.; 730 μm tip i.d.) held at 1.5 kV. The ion transfer capillary was held at 220°C. Full mass spectra were recorded in positive ion mode over a 400–1600 m/z range and with a resolution setting of 70000 FWHM (@ m/z 200) with 1 microscan per sec. Each full scan was followed by 7 MS/MS events, acquired at a resolution of 17,500 FWHM, sequentially generated in a data dependent manner on the top seven most abundant isotope patterns with charge ≥2, selected with an isolation window of 2 m/z for the survey scan, fragmented by higher energy collisional dissociation (HCD) with normalized collision energies of 30 and dynamically excluded for 30 s. The maximum ion injection times for the survey scan and the MS/MS scans were 50 and 200 ms and the ion target values were set at 106 and 105, respectively.
All data generated were searched using the Sequest search engine contained in the Thermo Scientific Proteome Discoverer software, version 1.4. The experimental MS/MS spectra were correlated to tryptic peptide sequences by comparison with the theoretical mass spectra obtained by in silico digestion of the Rattus norvegicus protein database downloaded January 2013 from the National Centre for Biotechnology Information (NCBI) website (www.ncbi.nlm.nih.gov). The following criteria were used for the identification of peptide sequences and related proteins: trypsin as enzyme; three missed cleavages per peptide were allowed and mass tolerances of ± 50 ppm for precursor ions and ± 0.8 Da for fragment ions were used. Validation based on separate target and decoy searches and subsequent calculation of classical score-based false discovery rates (FDR) were used for assessing the statistical significance of the identifications. Finally, to assign a final score to proteins, the SEQUEST output data were filtered as follows: 1,5; 2.0; 2.25 and 2.5 were chosen as minimum values of correlation score (Xcorr) for single-; double-; triple- and quadrupole-charged ions, respectively. A high stringency was guaranteed using parameters previously described and the false-positive peptide ratio, calculated through a reverse database, was less than 3%. The output data, protein lists, obtained from the Sequest search were compared using an in-house software, namely, the Multidimensional Algorithm Protein Map (MAProMa).
Normalized spectral abundance factor of the identified urine exosome proteins
The relative abundance of polypeptides was calculated from the normalized spectral abundance factor (NSAF) using the method of Paoletti et al taking into consideration the number of peptides as well as the length of the polypeptide contributing to their respective abundance. To enable comparison of samples from a subject across different time points or across different groups, each animal’s total proteome was normalized to 1. Subsequently the relative contribution of each protein from a given animal from a group was expressed as percentage of the total. Peptide numbers corresponding to a protein were thus more of raw data nature whereas the NSAF number included the peptide number as well as the total length of the protein. The peptide counts data were log-transformed prior to analysis by multivariate partial least squares discriminant analysis (PLSDA), and univariate 1-way ANOVA with unpaired comparisons, Variable Importance in Projection analysis and post hoc correction by Wilcoxon Rank test in MetaboAnalyst.
Data processing and statistical analysis
In all analyses, p ≤ 0.05 was considered statistically significant. Analyses including Student’s t-tests, Partial-Least Squares Discriminant Analysis (PLS-DA) and variable importance in projection (VIP) were performed with the MetaboAnalyst 2.0 web portal (www.metaboanalyst.ca). To reduce systematic variance and to improve the performance for downstream statistical analysis normalization and transformation of raw data were performed before the t-tests, PLS-DA and VIP analysis. Normalization by sum of the spectral count as mentioned previously was used to overcome the variance between the analyzed samples. To make each feature comparable in magnitude to each other, data were transformed by taking the natural log of the concentration values of the analyzed proteins. The data were additionally auto-scaled (mean-centered and divided by the standard deviation of each variable).
Univariate analysis was used to check the differences in the concentrations of the analyzed exosome protein spectral count between the control and infected rat urine samples. The infected rat urines were also split into male rat and female rat categories. Paired Student’s t-test was applied to examine each variable (ratio of individual protein to total concentration in each group considered).
PLS-DA and VIP were used both for the classification and significant feature selection. A VIP plot, which is commonly used in PLS-DA, ranks proteins based on their importance in discrimination between the urinary exosomes from infected and the uninfected rats. The VIP score is a weighted sum of squares of the PLS loadings. The amount of explained Y-variance in each dimension influenced the weights. Protein candidates with a false discovery rate (FDR) of ≤10% were qualified for subsequent validation by Western immunoblotting.
Western immunoblotting and quantification
Antibody against alanyl aminopeptidase was purchased from Proteintech Group, Inc., (Chicago, IL, USA). The THP antibody was from Sigma Chemical Co (St. Louis, MO, USA). HRP-conjugated secondary antibody was from GE Life Sciences (Piscataway, NJ, USA). SDS-PAGE gels (with 10% acrylamide) were used to resolve 100 μg of protein either from exosomes or total urines of male and female rats infected with L. interrogans serovar Copenhageni. After separation the proteins were transferred to nitrocellulose paper, blocked, and incubated with primary antibody overnight before washing with Tris-buffered saline, incubation for 1 h with HRP-secondary antibody conjugate and visualized by developing as described in previous publications from our laboratory. The quantification of the Western immunoblot bands was performed using Image J software (NIH) as previously described, and plotted using Graphpad Prism software (San Diego, CA, USA).
Urine protein content is different in Lepto-infected and uninfected control rats
Rat urine samples were analyzed for overall protein identification by a combination of SDS-PAGE and mass-spectrometry. We found that the infected rat urine shows an overwhelming increase in both quality and quantity of the protein content, as shown in Panel A of Fig. 1 (156 versus 503 proteins unique to uninfected versus infected urines). A total of 842 proteins were detected, with further classification of subgroups in the infected animals based on gender as shown in Panel B of Fig. 1. In total, 842 proteins were identified from the total urine animals with distribution as shown in Fig. 1 Panel A and B). Importantly, 180 proteins and 272 proteins were unique to the urines of female and male rats infected with L. interrogans serovar Copenhageni respectively, as compared to 156 proteins unique to the uninfected rat urines. These differences in the composition and quantity of proteins from infected and uninfected rats potentially indicates the reactions induced in these animals by leptospiral infection. The identity of each of these proteins is as given in S1 Table.
Urine exosomes from Leptospira-colonized rats show different protein constitution compared to uninfected control rats
Given that the urine exosomes reflect intracellular milieu of all types of various cells lining the nephron in kidney, and emerging evidence from literature that kidney functional alterations are induced due to leptospiral infection, we next focused on exosomes. We found that a total of 204 exosome proteins were identified classifiable into 7 different groups as noted in S2a-S2g Table and summarized in the Venn diagram (Panel C, Fig. 1). The exosome protein constitution also showed increase in the number of proteins expressed in the exosome, viz 32 proteins uniquely present in the uninfected exosome versus 57 unique proteins in the infected rat exosome.
We further conducted the multivariate partial least squares-discriminant analysis (PLS-DA) on these proteins. The analysis depicted in Fig. 2 shows clear separation between uninfected and infected rat urine exosome protein content, suggesting the differences between uninfected and infected rats.
Sex-specific alteration of the rat urine exosome protein content in relation to leptospiral colonization
Urinary exosome proteins in male vs. female infected rat urine were different as determined by PLS-discriminant analysis (Fig. 3). Moreover, the infected male rat urine exosome contents were far more different from those of uninfected rats compared to the infected female rat urine exosome contents. The VIP (Variable Importance in Projection) score of 25 proteins was higher than 1.5 (Fig. 4 and Table 1). Qualitatively, a total of 57 proteins were present among all the infected rats. Of these, only 3 were shared between infected males and females, while 37 were unique to infected males and 17 were unique to infected females. Further, we conducted separate analyses of proteins between proteins of proteins of male infected and female infected rats. Table 2 depicts male infected versus control rat urine exosome analysis. Accordingly, 11 proteins were significantly altered (p < 0.05). Table 3 depicts female infected versus control rat urine exoosme analysis, according to which the number of significantly dysregulated proteins was 9. In the male infected rat urine exosome, the alanyl (membrane) aminopeptidase upregulation not only reached the highest level of significance (p = 0.00019) but also had the lower FDR (3.22%). In the female infected rat however, although this upregulation was significant compared to the uninfected rat, the FDR did not reach the cutoff of <10% (14.37%). Thus both qualitatively and quantitatively, the protein content of exosomes showed gender specificity in infected rats.
Of interest, the top discriminator between control and infected rats, namely alanyl aminopeptidase of the membrane origin,is also known as aminopeptidase neutral (APN) or CD13. Furthermore, CD13 also shows different levels of dysregulation between infected male and infected female rats (Table 1).
The urinary exosome membrane alanyl aminopeptidase or CD13 is a marker of infection
A separate analysis of the proteins dysregulated between control and infected rats showed that 11 proteins were significantly different in urinary exosomes (p <0.05, Table 4). However, only one protein had an FDR of <10%, namely alanyl (membrane) aminopeptidase, also known as CD13. By both multivariate analysis of PLS-DA (Fig. 4) and univariate ANOVA, CD13 is significantly upregulated in the infected rat urine exosome. When the analysis was conducted without gender specificity (Table 4), only CD13 showed an FDR of <10% among these 11 dysregulated proteins. Taking into account gender specificity (Table 2) CD13 was not only significantly upregulated, but also had an FDR of <10%. However, in females, this protein was only significantly upregulated but did not reach the FDR cutoff (Table 3).
The VIP scores of proteins identified in the rat urine exosomes shows CD13 to be a top discriminant between infected and uninfected rat urine exosome (Table 1). Twenty-five proteins had a VIP score of >1.5, fulfilling criteria to classify a protein as a reliable discriminant. Our analysis shows that this value 5.72 for CD13.
Further, Western immunoblotting of the protein showed that the exosome content of CD13 closely tracked the proteomic data, with a slight increase in the infected female urine exosome, and robust increase in the infected male urine exosome (Fig. 5a). The observed difference was significant (Fig. 5b).
Tamm-Horsfall Protein is significantly decreased in the infected rat urine
Given the tubular location of CD13, we tested the hypothesis that other proteins reflecting the tubular function may be affected in response to leptospiral infection. We chose to study the most abundant protein in the normal urine, namely the Tamm-Horsfall Protein (THP). Western immunoblotting of THP in urine of infected and uninfected rats showed robust THP expression in the uninfected rats and significantly lower THP in infected male and female rats, supporting this hypothesis (Fig. 6).
Conclusions and future directions
Many conclusions can be drawn from this study:
Although the general quantity of proteins in total urine and exosome component show a similar trend, qualitatively the proteins are different based on identifications, the GI numbers and hence the pathways they belong to.Utility of exosome Analysis in characterizing a complex biochemical phenomenon such as the host-response to leptospirosis infection is very high, and has many implications:
Public health interest: given that nephron-cell specific cargo that urine exosome carries, primarily the pathway employed or dysregulated for infecting the host to bring about the infection phenotype, and secondly, the response that is unique to each host studied potentially opens up the possibilities of disease-specific and severity-specific treatment to leptospirosis.Non-invasive: it would be ideal to biopsy every infected individual to understand the various facets of leptospirosis infection in order that our knowledge about Leptospira increases. However, it is neither practical nor ethical, given the invasive nature. Urine exosome analyses is non-invasive, highly specific and provides a window into the organ level structural or functional changes. If implemented quickly enough in a translational research setting, this specificity potentially imparts ability to move the field towards personalized medicine.Gender-specific host responses to disease onset: our data shows that this methodology of studying differences between male and female host response to infection can be applied in the setting of human disease. Although we should expect different molecules other than CD13 to by dysregulated, the individual-specific pathways mediating different magnitude responses to the same infection can be accurately measured using exosome markers that are either surrogately or directly linked to a particular host-response event. This assumes importance especially in the setting of sub-clinical leptospirosis infection where in the patient does not display any clinical features/symptoms but continues to play host to the bacterium.