Reagents

Restriction enzymes Xba I, Hind III and T4 ligase were purchased from Fermentas (Thermo Fisher Scientific, Germany). DNA markers and protein markers were purchased from LabLead Co., Ltd (Beijing, China). PCR mix was purchased from JiErDun Co., Ltd (Shanghai, China). Salmon sperm DNA, hemin, and TMB were purchased from TaiTianHe Co., Ltd (Jinan, China). A turbidimetric (1,3)-β-D-glucan test kit was purchased from Bokang Marine Biological Company (Zhanjiang, China). Barley glucan, curdlan, mannan, endotoxin, dextran, and (1,3)-β-D-glucanase were sourced from Sigma–Aldrich (St. Louis, MO). ssDNA libraries and primers were synthesized by Sangon Biotech Co., Ltd (Shanghai China).

Extraction of (1 → 3)-β-D-glucans

The strain (C. albicans ATCC 10,231) was purchased from the China Center for Type Culture Collection affiliated with Wuhan University. This strain was cultured on Sabaurauds agar (SAB) medium and incubated (48 h at 34 °C).

The (1 → 3)-β-D-glucans were isolated from C. albicans, as previously described [29]. The extraction was done as follows, lane 1: The extraction of rude (1 → 3)-β-D-glucans from Candida albicans with Chemical method and ultrasonic crushing; lane 2:Deionized water blank control; lane 3 to lane 6: The rude glucans was digested with (1,3)-β-D-glucanase in different times,1 h, 1.5 h, 2 h and 2.5 h respectively. After treatment, Candida albicans (1,3)-β-dextran, analyzed with silver nitrate staining and periodic acid sif staining, respectively. It was then digested with (1 → 3)-β-D-glucans (Sigma Aldrich) to give a less polymerized water-soluble (1 → 3)-β-D-glucans (< 1.7 KD) as the target of aptamer screening. Low degree polymerization (1 → 3)-β-D-glucans were separated using 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and identified via periodic acid–Schiff (PAS) staining. Next, The purified low degree of polymerization (1,3)-β-d-dextran was analyzed by high performance liquid chromatography and elemental analysis.

In-vitro identification of anti-(1 → 3)-β-D-glucans aptamers

Aptamers with high and specific binding affinity to (1 → 3)-β-D-glucans from C. albicans were screened, as previously described [28, 29]. A random ssDNA aptamer library (1014 ~ 1015) was synthesized, containing 50 random nucleotides (50N), and flanked by fixed sequences (the complementary fragments of the G-quadruplex): 5′-TCTAGAATCCCAATCCCAATCCCA-50N-ACCCTAAAGCTT-3′ (86 nucleotides, about 3.4KD; “N” denotes any of the bases A, T, G, and C) was used as the initial pool. The underlined fragments refer to the cutting sites of the restriction enzymes Xba I and Hind III, respectively. The italicized fragments correspond to the reverse sequences of the G-quadruplex. Primer 1 (P1: 5′-TCTAGAATCCCAATCCCAATCCCA-3′) and primer 2 (P2: 5′-AAGCTTTAGGGT-3′) were used for PCR-based amplification of the DNA library.

The selection steps were as follows: first, 50 μL of purified (1,3)-β-D-glucan (10 μg/mL) were coated on polystyrene microwells with carbonate buffer (pH = 9.6) for 1 h at 37 °C; subsequently, wells were blocked at ambient temperature (25 °C) for 1 h using salmon sperm DNA. Before adding to the coated wells, the initial ssDNA library was suspended in screening buffer (136.9 mM NaCl, 3.98 mM MgSO4, 2.68 mM KCl, 1.8 mM CaCl2, 1.47 mM KH2PO4, and 8.06 mM Na2HPO4), heated (10 min at 95 °C), and instantly placed on ice (5 min). Subsequently, 1 nmol initial ssDNA library was added to the coated wells and incubated (1 h) in a water bath (37 °C). The liquid in wells was discarded and the wells were washed with buffer (screening buffer containing 0.1% Tween-20) three times. Thereafter, 100 μL of deionized H2O was added into coated wells, and the wells were heated (95 °C for 10 min). The eluted ssDNA was removed from the wells and used as the template for asymmetric PCR amplification for next screening round. These PCR reactions were performed in 50 μL total volumes, containing 2 μL of template DNA, 25 μL of PCR mixture (2×), 1 μL of each primer (10 M), and 24 μL of nucleic acid-free distilled water on a Bio-Rad PCR System (T100, Bio-Rad, USA). PCR conditions included a pre-warming step of 5 min at 95 °C, 12 cycles performed; 95 °C for 30 s, 60 °C for 30 s, and 72 °C for 20 s, a final extension step at 72 °C for 10 min followed. The complete in-vitro selection required a total of eight selection cycles.

Finally, we selected the highest affinity pool among the eight pools and amplified the ssDNA of the highest affinity pool to yield double-stranded DNA (dsDNA) by routine PCR. The dsDNA and the pUC19 plasmid were cut by Xba I and Hind III and ligated using a T4 ligase. The recombinant plasmids were transferred into E. coli DH5α, and 40 single aptamer clones were randomly picked and amplified by PCR for identification. Plasmids from 40 positive clones were PCR-amplified with a biotin-labeled primer for generation of biotinylated single ssDNA aptamers for subsequent evaluation of their affinity to (1 → 3)-β-D-glucans by enzyme-linked oligonucleotide assay (ELONA).

Binding affinity and specificity of selected individual aptamers determined by ELONA

For testing the binding affinity, 96-well plates were treated with 50 μL of purified (1,3)-β-D-glucan (10 μg/mL) and incubated overnight (4 °C), washed thrice with phosphate-buffered saline (PBS), then blocked for 1 h at 37 °C with salmon sperm DNA (100 μL, 10 μg/mL). The different types of (1 → 3)-β-D-glucans (curdlan and barley glucan) and non-(1 → 3)-β-D-glucans (mannan, endotoxin, and dextran) were solubilized according to the suppliers’ instructions. These polysaccharides (10 μg/mL) were coated on 96-well plates and left at 4 °C overnight, rinsed three times with PBS, and blocked with salmon sperm DNA (100 μL, 10 μg/mL; 1 h at 37 °C).

Biotin-labeled aptamers (10 nM) were then added to the above-described microwells coated with different substances and incubated (1 h at 37 °C). Post rinsing the wells with PBS (five times) to remove excess unbound aptamers, HRP-conjugated streptavidin was added and left to incubate (37 °C for 30 min). Subsequently, the wells were washed three times using PBS. Finally, TMB substrate and stop buffer (2 M H2SO4) were added and absorbance values were measured at optical density (OD)450 nm using a microplate reader.

Development of the aptamer G-quadruplex/hemin self-assembling color system (AGSCS)

Equal amounts of aptamers (a mixture of four aptamers [A1, A4, A5, and A6]) and G-quadruplex were mixed to combine G-quadruplexes with their complementary sequences in the tail of the aptamers (Scheme 1a). In the presence of (1 → 3)-β-D-glucans, aptamers specifically recognized and bound to (1 → 3)-β-D-glucans, the hairpin structure of the aptamer probe was opened, and the G-quadruplex was released, resulting in the formation of a G-quadruplex/hemin complex (Scheme 1b) with the aid of hemin, which exhibited biocatalytic abilities in the process of H2O2 electroreduction and generating an amplification signal for (1,3)-β-D-glucan detection. However, in the absence of (1 → 3)-β-D-glucans, the G-quadruplex binds to complementary sequences in the aptamer tail to form a short double-stranded DNA (dsDNA). Hence, the G-quadruplex would not be released from the aptamer, thus the G-quadruplex/hemin would not be formed in presence of hemin, and the reaction with Scheme 1 and 2 TMB and H2O2 to generate a visible color change for (1 → 3)-β-D-glucans detection would not be catalyzed (Scheme 1c).

Scheme 1.

Schematic diagram of the elements and working process of the aptamer G-quadruplex/hemin self-assembling color system (AGSCS). A Diagram of the structural composition of the aptamer with G-quadruplex. The blue non-closed ring represents the aptamer; the black bases reflect the complementary sequences of the G-quadruplex; the red bases represent the G-quadruplex sequences. B Diagram of the structural composition of the G-quadruplex/hemin complex. The “cube” comprises 12 guanines from the same G-quadruplex line, with every four guanines distributed in a plane. The green hexagon represents hemin. C The working process of the AGSCS. The aptamers with G-quadruplex were added to the samples and mixed carefully. In the presence of (1 → 3)-β-D-glucans, aptamers bound the target tightly, the hairpin structure of the aptamer probe was opened, and the G-quadruplex was released. The free G-quadruplex was folded into a cube containing hemin with the aid of hemin in the system. The G-quadruplex/hemin complex showed peroxidase-catalytic activity and catalyzed the tetramethylbenzidine/H2O2 system to change color, from colorless to blue. After the addition of stop solution, there was a color change from blue to yellow, and the color could be detected at 450 nm by a microplate reader or the naked eye

Scheme 2

Schematic diagram of results scored by the naked eye with color card. A To score the results via naked eye, a color card was designed as a tool for enabling accurate and objective diagnosis. In this card, six different patterns in the same color, were distributed in a row as indicated in the diagram. B The pattern was covered by tubes containing different shades yellow, from tube 1 to tube 5. If the pattern could be distinguished by the naked eye [e.g., as is the case with tubes 1, 2, and 3 in B, the tube containing the sample should be scored as negative, and the patient would not be categorized as having invasive fungal infections (IFI). If no pattern was observed (e.g., as is the case with tube 5), the tube containing the sample should be scored as positive, and the patient would be considered to have IFI. If the pattern could be distinguished faintly (e.g., tube 4), the tube containing sample should be scored as weakly positive, and the patient could potentially have IFI

Quantitative detection of (1,3)-β-D-glucan by the AGSCS

Different concentrations of (1,3)-β-D-glucan from 0 to 400 pg/mL (400 pg/mL, 200 pg/mL, 100 pg/mL, 50 pg/mL, 25 pg/mL, 12.5 pg/mL, 6.25 pg/mL, 3.125 pg/mL, 1.6 pg/mL, 0.8 pg/mL, and 0 pg/mL) were used to coat 96-well plates (50 μL/well), which were kept overnight (4 °C), washed with PBS (three times), and blocked with salmon sperm DNA (100 μL, 10 μg/mL) for 1 h at 37 °C. Subsequently, 100-nM aptamers (a mixture of four aptamers [A1, A4, A5, and A6]) and 100-nM G-quadruplex were mixed to make G-quadruplexes combine with their complementary sequences in the tail of aptamers. Aptamers with G-quadruplex were added to the above-mentioned wells, and hemin was added at the same time. Plates were incubated (1 h at 37 °C) to form the AGSCS. After adding TMB/H2O2 mixture substrate for coloration and 2 M H2SO4 stop buffer, absorbance values were measured at OD450 nm using a microplate reader.

Participants

Serum samples from 82 patients with IFI (including 14 patients with a central venous catheter, eight patients subject to thoracic surgery, 10 patients with severe burns, 27 patients with cancer, and 23 patients with chronic respiratory failure) were taken from Second People’s Hospital of Wuhan, Zoucheng People’s Hospital, and the Second Affiliated Hospital of Wannan Medical College from February 2016 to February 2019. The patients had an average age of 56.2 years, including 45 males and 37 females. The inclusion criteria were as follows: (1) positive fungal culture or body fluid smear; (2) symptoms and signs of fungal infection (e.g., hypotension and/or fever); (3) documented curative effects of anti-fungal therapy; and (4) positive G test and procalcitonin test. All studied participants were required to meet at least three of these criteria. Serum samples from 116 healthy individuals were collected from the Physical Examination Center of the Second Affiliated Hospital of Wannan Medical College and the Second People’s Hospital of Wuhu. The ethics committee of Second Affiliated Hospital of Wannan and other three hospitals mentioned above approved the study. All participants gave written informed consent.

(1,3)-β-D-glucan detection in human serum samples using AGSCS

The AGSCS procedure was carried out as follows: at first, a 96-well enzyme-linked immunosorbent assay (ELISA) plate was irradiated with ultraviolet light (2 h) and blocked using salmon sperm DNA (100 μL, 10 μg/mL; 1 h at 37 °C). A 50 μL volume serum sample, was added to the wells, and at the same time, aptamers with G-quadruplex and hemin were added, and plates were incubated (37 °C for 1 h). Finally, a TMB substrate together with the stop buffer were added to the wells. A microplate reader was used to measure absorbance at OD450 nm. Given our research purpose to develop a simple test, the AGSCS test results were also read by the naked eye. A simple and objective criterion was introduced for scoring the samples as “negative,” “positive,” or “slightly positive,” as indicated and explained in Scheme 2.

G test (commercial)

Quantification of (1,3)-β-D-glucan in 198 clinical serum samples was done using a turbidimetric (1,3)-β-D-glucan test kit (Cat#KT-120, Zhanjiang A & C biological Ltd. China) following the manufacturer’s protocol. For positive sample classification, the concentration of (1,3)-β-D-glucan needed to be ≥ 100 pg/mL. All positive samples were subject to repeated testing and considered positive only if the repeat result was also positive.

Guinea pig model

Male guinea pigs (weighing approximately 0.5 kg) were purchased from the Laboratory Animal Center of Wuhan University (Wuhan, China). Ten guinea pigs were injected subcutaneously with triamcinolone acetonide (Sigma-Aldrich) at 300 mg/kg body weight and immunosuppressed twice daily. On day 4, Candida albicans (5 107 CFU) is given intravenously in the saphenous vein, followed by triamcinolone acetonide reduced to 150 mg/kg once daily. Serum (1 3)-β-d-glucan levels are measured every 2 days with G-test and AGSCS.

Statistical analysis

Quantitative data were reported as mean ± standard deviations (SD) from three replications. Data analysis was done using IBM SPSS (version 25.0 for Windows (IBM, Armonk, NY, USA). One-way ANOVA with Student’s t-test was used to compare means. The chi-square test was used to compute the deviations between categorical variables. The probability, P < 0.05 were classified as statistically significant.