Introduction

Nowadays glycoconjugates represent an important class of efficient and safe glycoconjugate vaccines. The conjugation of a polysaccharide antigen with a protein or peptide via covalent linkage is one of the main recent trends to overcome shortcomings of old polysaccharide vaccines, such as a T-cell independence. However, while the traditional mechanism of action of glycoconjugates considered peptides generated from the carrier protein to be responsible for T-cell help recruitment, only recently the evidence of the active involvement of the carbohydrate part in determining the T-cell help has been shown [1]. Moreover, it was identified using radiolabelled glycoconjugates the means by how glycoconjugates are processed by B-cells [2]. For example, fluorescently labelled pneumococcal polysaccharides were used to determine the location of the polysaccharide component of the glycoconjugate in the antigen presenting cells (APCs) [3]. In another example radiolabelled (3H) group B streptococcal type III polysaccharide coupled to a carrier protein was used for processing by APCs, resulting in presentation of carbohydrate epitopes that were able to stimulate CD4+ T cells [2]. Thus, labelled glycoconjugates represent effective tools for tracking antigen exposure in immunobiological studies. However the involvement of intracellular vesicles in the trafficking of Candida mannan antigen is not known. The fluorescently labelled mannan is a useful tool for the investigation of uptake of the Candida albicans mannan and two-component conjugate systems (mannan as Candida saccharide antigen and Candida specific peptide antigen) and could effectively serve for intracellular trafficking by the APCs.

Our goal was to prepare a series of fluorescently labelled mannan glycoconjugates, namely mannan–peptide–label and mannan–label with well-defined composition as well as proportionally identical composition of saccharide, peptide and the fluorescent label.

In the presented work, fluorescently labelled mannan and mannan–peptide conjugates were prepared and characterized by FTIR-ATR spectroscopy, HPSEC and UV-VIS colorimetry. The cell penetration and transport of conjugates was successfully monitored using fluorescence microscopy imaging. We suppose that proposed synthesis of glycoconjugates is versatile, simple and can be used in various phases of artificial vaccine development and related fields. This is especially useful where whole protein conjugation is uncomfortable and where the fate of glycoconjugate in vitro is desirable to track.

Materials and methods

Chemicals used were: 2-aminobenzamide (anthranilamide), NaBH3CN, NABH4, DMP (Dess–Martin periodinane, 1,1,1-triacetoxy-1,1-dihydro-1,2-benziodoxol-3(1 H)-one), dry DMSO, fluorescein isothiocyanate (all from Sigma–Aldrich, Slovakia), DMSO (up to 0.5 vol.% water, Slavus, Slovakia), dextran (Biotika, Slovakia), peptide, H-Gln-Gly-Glu-Glu-Ala-Leu-Ile-Gln-Lys-Arg-Ser-Tyr-Lys-Lys-OH (M = 1908.16, prepared by KJ Ross-Petersen, Klampenborg, Denmark).

Isolation and ultrasound degradation of Candida albicans high-molecular-mass mannan

Cellular mannan was isolated and purified using Fehling’s reagent according to literature [4] from the yeast cell wall of Candida albicans serotype A CCY 29-3-100 (Culture Collection of Yeasts, Institute of Chemistry, Center for Glycomics, Slovak Academy of Sciences, Bratislava, Slovakia) as a high-molecular-mass polysaccharide.

Ultrasonic degradation was performed using BRANSON 450A Sonifier Analog Cell Disruptor (20 kHz; BRANSON Ultrasonics Corporation, Danbury, CT, USA) equipped with 19 mm high grain titanium probe as was described [5]. Briefly, 30 mL of aqueous solution of high-molecular-mass mannan (33.3 mg mL−1) was ultrasonicated at 60 W ultrasonic power for a total of 80 min. The temperature of the sample was kept below 50 °C during the ultrasonic treatment using ice-bath and by applying batch exposure to ultrasound. Any bits of titanium shed by the probe because of long exposure times were removed by centrifugation prior to further analysis. Solution of degraded mannan was freeze-dried and the product (mannan) was obtained in a form of white solid (M p = 35 kDa; M w = 44 kDa; M n  = 27 kDa; PDI = 1.6).

Dess–Martin periodinane oxidation

Typical small scale procedure: 5 mL of dry DMSO (Sigma–Aldrich, Slovakia) was added to 50.0 mg of mannan (n(monosaccharide unit) = 3.1 × 10−4 mol). The suspension was gently stirred until fully dissolved (≈ 3 h). Then, 1 eq. of Dess–Martin periodinane (DMP, 1,1,1-triacetoxy-1,1-dihydro-1,2-benziodoxol-3(1 H)-one) was added to the stirred mannan solution at 30 °C. The reaction was stopped after 1 h by adding 3 mL of ice-cold water. Precipitate was removed and supernatant was dialyzed (with 12,000 MWCO) as follows: 3 times against 0.1 % NaCl by mass, 0.1 % NaHCO3 solution and finally 6 times against pure water and then freeze-dried. Yield: 30 mg of oxidized mannan (60 %) with molecular mass of 26 kDa (72 % of the mannan), degree of oxidation (DO) was 10.9 %.

Preparation procedure: 82 mL of dry DMSO (water content ≤0.5 mass %) was added to 820 mg of mannan. The suspension was gently stirred to become fully dissolved (≈ 3 h). Then, 2.15 g (1 eq.) of Dess–Martin periodinane was added to the stirred mannan solution at 35 °C. The reaction was stopped after 10 min by adding 160 mL of cold water. Precipitate was removed and supernatant was dialyzed (with 12,000 MWCO) as follows: 2 times against 0.1 mass % NaCl and 0.1 mass % NaHCO3 solution (4.5 L) and finally 5 times against water (4.5 L) and then freeze-dried. Yield: 733 mg of oxidized mannan (89 %) with molar mass of 36 kDa, degree of oxidation (DO) was 8.9 %.

Preparation of labelled mannan and mannan–peptide conjugate

A water solution of oxidized mannan (OM) (1.00 mL; 15 mg mL−1) was prepared and peptide solution in phosphate buffer (pH 7.0) was added and stirred for 30 min at (26.0 ± 1.0) °C. Next, fresh solution of 0.34 mg NaBH3CN in 0.05 mL PBS (pH 7.0, 0.05 M) was added. The reaction progress was monitored as CHO content decreased in relation to OM content [6]. After 4 h the reaction was completed. At this point, the reaction was quenched or used for further labelling with 2AB. In first case the reaction was quenched with NaBH4 and stirred for 2 h followed by dialysis and freeze-dried. Yield was 12.3 mg of mannan–peptide (82 %).

Fresh solution of 0.355 mg of 2AB in 0.05 mL of acetate buffer (pH 4.5, 1 M) was added to a solution of OM (15 mg mL−1) or mannan–peptide (reaction mixture from above procedure). Then, a fresh solution of 0.34 mg NaBH3CN in 0.05 mL of the acetate buffer was added and the reaction progress was monitored by TLC. After 30 min there were still traces of unreacted 2AB therefore another 0.05 mL of NaBH3CN solution was added. The reaction was completed in next 30 min and quenched by addition of NaBH4 and stirred 2 h. 3 mL of water was added and dialyzed against water (4 × 0.75 L, 12,000 MWCO) or purified in centrifugal sieves Amicon (0.5 mL, 3000 MWCO) and then freeze-dried. Yields were 12.7 mg of mannan–2AB (85 %) and 14.1 mg of mannan–peptide–2AB (94 %).

Recovered mannan was prepared from solution of OM (15 mg mL−1) by reduction with NaBH4. The reaction mixture was monitored and treated as above. Yield was 11.1 mg of recovered mannan (74 %).

Analytical

High performance liquid chromatography (HPLC) using two HEMA-BIO 100 and 300 columns (8 mm × 250 mm) connected in series was used to characterize the prepared constructs. The mobile phase used was 0.1 M NaNO3. The carbohydrate content in the eluent was monitored by differential refractometer RIDK-102 (Laboratorní přístroje Praha, Czech Republic).

A set of dextrans (American Polymer Standard Corporation, Mentor, OH, USA) were used for calibration of the HPLC system.

The fluorescently labelled samples were further analysed on Accela UHPLC system (Thermo Fisher Scientific Inc.; Waltham, MA, USA) equipped with Accela 1250 Pump and Thermo Scientific Dionex UltiMate™ 3000 fluorescence detector FLD-3100 operating at excitation wavelength of 319 nm and emission wavelength of 419 nm. Two HEMA-BIO 100 and 300 columns (8 mm × 250 mm) connected in series were used for analysis. The mobile phase used was 0.1 M NaNO3.

FT-IR spectra were measured on NICOLET Magna 6700 (Thermo Fisher Scientific, USA) spectrometer with DTGS detector, experimental accessory – Smart Orbit and OMNIC 8.0 software was used. Infrared spectral analyses were carried out in mid-infrared region (from 4000 cm−1 to 400 cm−1) and spectral data obtained were presented as absorbance values. Number of scans was set to 128. Diamond Smart Orbit ATR accessory was used for measurements in solid state.

UV-VIS spectroscopy: The 2AB or tyrosine content in mannan derivates was estimated by measurement of absorption at 311 nm or 280 nm, respectively, on Shimadzu UVmini-1240 spectrometer. 2AB or tyrosine was used as a standard, respectively. Results were expressed as a mean value with standard deviation for n measurements.

Determination of C==O groups: The degree of oxidation of polysaccharides was determined by Park–Johnson colorimetric assay. The determination of the contents of carbonyl groups is based upon the reduction of ferricyanide ions in alkaline solution [6]. Mannose and glucose were used as standards. The results were expressed as % of oxidized polysaccharide monomer unit.

Thin-layer chromatography was performed on aluminium sheets with silica gel 60 F254. Mobile phase used was CHCl3–MeOH–AcOH–H2O (60: 30: 3: 5, vol.). Fluorescent analytes were detected under UV light: R(2AB) = 0.9, R(mannan–2AB) = 0.

Prior to NMR spectroscopy, the samples were exchanged twice with D2O and then dissolved in 99.996 % D2O. All NMR spectra were recorded on Varian VNMRS 600 MHz spectrometer equipped with cryogenically cooled inverse HCN probe at 25 °C. 1 H–13C HSQC spectra were acquired using adiabatic pulses for inversion and referenced to internal acetone (2.225, 31.07 ppm).

Cell cultivation and activation

RAW 264.7 cell line murine macrophages (ATCC®TIB-71™, UK) were cultured in complete DMEM (Sigma, USA) for 24 h (until approx. 80 % confluent). Subsequently, the aliquots of 1 × 105 cells mL−1 per well (with 93.1 % viability) were seeded in triplicates into 24-well cell culture plate (Sigma, USA), and exposed to 0.6 mg of mannan–2AB or mannan–peptide–2AB conjugate. In vitro exposition was performed for 72 h at 37 °C in an atmosphere of 5 % CO2 and at 90–100 % relative humidity. Cell viability was assessed using the trypan blue exclusion method.

Immunocytometry

Fluorescein isothiocyanate (FITC) labelled mannan was prepared in order to quantitatively measure the incorporated fluorescence in mice macrophages cell line RAW 264.7 as APCs by immuno-flow cytometry. 2 μL of pyridine and 18 mg of FITC were added to the solution of 54 mg of mannan in 1 mL of DMSO. Reaction mixture was heated to 95 °C for 2 h. Afterwards 10 mL of water was added. After extensive dialysis and freeze-drying, 29 mg of product with degree of substitution of (1.2 ± 0.1) % was obtained.

The 100-μL aliquots of RAW 264.7 cells following incubation (2 h, 4 h, 20 h and 24 h) with 0.25 mg, 0.125 mg and 0.025 mg of mannan–FITC/105 cells were subjected to immuno-flow cytometry using a Beckman Coulter FC 500 flow cytometer (Beckman Coulter, Fullerton, CA, USA) equipped with a 488-nm argon laser and a 637-nm HeNe collinear laser and controlled by CXP software. RAW 264.7 cells were gated on the basis of forward light scatter (FSC) and side light scatter (SSC), using a linear scale. Gates were set to exclude the debris and damaged cells using forward scatter vs. side scatter dot plot discrimination and settings were optimized with properly prepared Mannan–FITC untreated cell culture. For each sample fluorescence histograms of 10 000 cells were generated and analysed (green fluorescence, 525-nm band-pass filter, FL1 channel). All samples were analysed in duplicates. The data are expressed as a mean of fluorescence intensity, MFI ± SD.

Cell surface fluorescence quenching

To quench the extracellular fluorescence, 0.4 mass % trypan blue (Sigma–Aldrich, USA) was used. Immuno-cytometric analysis of trypan blue treated RAW 264.7 cells was performed till 30 min. Using the same protocol as previously described.

Fluorescence microscopy

The intracellular internalisation of mannan–2AB, mannan–peptide–2AB and mannan-FITC was evaluated based on visualizing of in situ fluorescence using fluorescence microscope Imager A.1, running under Axio-Vision software and equipped with AxioCam MRC camera and Plan-Neofluar objectives (Zeiss, Germany), magnification 630×, fluorescence filter set specific for 2-AB, FS 49 and for FITC, FS 09 (Carl Zeiss, Germany).

Results and discussion

Preparation of oxidized mannan

Mannans of Candida species are neutral surface high-molecular-mass polysaccharides, which could be easily isolated from the yeast cell walls. Candida yeast mannan consist of α (1–6)-linked mannose backbone with various frequency of branching. Branches contain α (1–2)-, α (1–3)- and β (1–2)-linked mannooligomers [7]. Mannan, among other polysaccharides, is intensively studied as a potential vaccine immunoactive atigen [8] or as a carrier for immunogens [9, 10]. High-molecular-mass cell-wall mannan from Candida albicans serotype A is a branched polysaccharide with a broad molecular mass distribution (PDI ≈ 3.3) and molecular masses of about 1 MDa. By means of ultrasonication, we were able to prepare a well-defined low-molecular-mass mannan (≈ 35 kDa) with relatively narrow molecular mass distribution as evidenced by decrease of PDI to 1.6 [5]. 1H NMR spectrum of the ultrasonicated mannan was the same as that of the original isolated mannan and 1 H–13C HSQC (not shown) spectrum corresponds to the one in literature [5]. The ultrasonicated mannan was not directly suitable for conjugation, because it did not have sufficiently reactive groups. Therefore, we (mildly) oxidized only the C6 position of mannose units to their aldehyde derivatives. For mannan oxidation we used Dess–Martin periodinane, which is known to selectively oxidize primary rather than secondary alcohols [11]. This mildly oxidized mannan can be restored back to original polysaccharide by mild reduction with sodium borohydride. Dextran 40 was used as a test compound [12, 13]. With the dextran we observed the degradation process along with oxidation reaction (Fig. S1a). The polymer chain degradation of dextran was also previously described [12, 13]. The aldehyde group formation was confirmed by FTIR (\( \tilde{\nu} \) = 1741 cm−1, Fig. S2a). The mannan oxidation at the same conditions resulted in almost no degradation and higher degree of oxidation (DO) (Fig. S1b). Finally, we prepared oxidized mannan (OM) by quenching the reaction after 10 min by addition of cold water. Estimated DO was 8.9 %, and M = 36 kDa. The FTIR spectrum confirmed the CHO group formation at \( \tilde{\nu} \) = 1738 cm−1, Fig. S2b. From the comparison of original and oxidized mannan 1H–13C HSQC spectra (Figs. S3a and S3b, respectively) it is evident that a new crosspeak appeared at 5.27/89.1 ppm. This crosspeak was identified as a hydrated aldehyde group (CHO) at C6 position: both proton and carbon chemical shifts for this CHO group are in agreement with manno-hexodialdo-1,5-pyranoside [14]. In order to confirm the presence of the hydrated aldehyde group, OM was reacted with hydroxylammonium sulfate, which converted the CHO group to aldoxime. Consequently, the crosspeak of the hydrated CHO group moved to 7.56/150.4 ppm as is exhibited in the HSQC spectrum (Fig. S3c). Again, the values of both proton and carbon chemical shifts for the aldoxime group are in agreement with previously published data [15].

HPSEC chromatograms indicate that OM in solution consists of two different species. The major component has the expected mass of 36 kDa and a minor one has a higher mass of 97 kDa, indicated by a shoulder of the major peak (Fig. 1a). The minor 97 kDa component is probably an aggregate of the major one. After reduction of aldehyde groups, we obtained the recovered mannan (RM), free of aldehyde groups, with only a slight decrease of molecular mass (33 kDa) compared to mannan (Fig. 1a). This observation suggests that the aggregates of OM were formed solely from molecules of OM and their formation is reversible.

Fig. 1
figure 1

HPSEC chromatograms (refractive index detection) of Mannan (M), Oxidized mannan (OM) and Recovered mannan (RM) (a); Mannan–2AB (M-2AB), Mannan–peptide–2AB (MP-2AB) and Mannan–peptide (MP) (b).

Full size image

Preparation of fluorescently labelled mannan

Recently, a number of fluorescent labels have been introduced and some of them were efficiently applied for the analysis of oligosaccharides [16]. To ensure reasonable results from analysis of the conjugate-processing by APCs (e.g. macrophages) in vitro the exact determination of conjugates composition is crucial. Aminobenzene derivatives substituted at the position 3 showed good reactivity with reducing carbohydrates, although the fluorescence intensity and molar absorptivity of these derivatives were not as high as those of 2- and 4-aminobenzene derivatives [17]. To achieve this intention, 2-aminobenzamide (2AB) was selected for several reasons. 2AB is a small molecule that reacts well with carbohydrates with excellent fluorescence intensity [18–20] and possesses low cytotoxic activity [21]. 2AB should be safe as a chemical, while not yet tested in vivo, it was assumed to have minimal immunogenicity and toxicity [22]. 2AB was also used as a linker to conjugate free reducing glycans with proteins. Moreover, antiproliferative activity accompanied with low toxicity of histone deacetylase inhibitor comprising of 2-aminobenzamide group was reported [23].

We investigated the utilization of available protocols for oxidized polysaccharide labelling [20]. Reductive amination reaction was conducted at 65 °C, with a large molar excess of 2AB to carbohydrate. Unexpectedly, we observed very high polysaccharide chain degradation. We found out that the reaction can proceed at ambient temperature and with addition of much lower portion of fluorescent amine. We added enough 2AB to the mannan with DO of 7.9 % to occupy 2.8 % of mannan residues (35 % of aldehyde groups). The reaction progress was monitored by TLC where the polysaccharide remained at the start, while the 2AB spot has moved (R = 0.9). 10 min after the NaBH3CN addition the spot of 2AB disappeared while spot of the polysaccharide at R ≈ 0 emitted light. At this point the reaction was quenched by addition of NaBH4 to reduce all remaining CHO groups.

The product (mannan–2AB) was investigated by HPSEC equipped with refractive index detector (RID) (Fig. 1b) and fluorescence detector (FLD) (Fig. 2a). Both chromatograms show single peak of the product with mass of 28.2 kDa. The UV spectroscopy confirmed the 2AB content to be (2.2 ± 0.1) mass % or (2.7 ± 0.1) mole %.

Fig. 2
figure 2

HPSEC chromatograms (fluorescence detection) of Mannan–2AB (M-2AB), Mannan–peptide–2AB (MP-2AB), λ exc = 319 nm, λ em = 419 nm (a); Mannan–peptide–2AB (MP-2AB) and Mannan–peptide (MP), λ exc = 274 nm, λ em = 303 nm (b)

Full size image

Preparation of mannan–peptide conjugate

A peptide was selected (consisting of 14 aminoacids) according to previously identified Candida cell wall proteins that are expressed during pathogenesis of human disseminated candidiasis [24]. The selected peptide represents a sequence located in Candida hyphal wall protein-1 (Hwp1). It shows 100 % homology only with C. albicans and does not have identity with mammalian proteins. After conjugation with β-mannan the trisaccharide (β-(Man)3) possess ability to induce protective anti-Candida immune response [24]. This design was adopted from literature to be related to pathogenesis of human candidiasis [24]. Furthermore, Lys-Lys terminal dipeptide was incorporated to enhance the peptide reactivity towards aldehyde groups.

Peptide and OM were coupled in the conditions of reductive amination similar to those of 2AB but at pH 7. The reaction progress was monitored indirectly, as a decrease of CHO content by the method of Park and Johnson [6]. Once the CHO content has been stabilized, the reaction mixture was quenched by addition of sodium tetraborohydride.

HPSEC RID (Fig. 1b) and FID (Fig. 2b) chromatograms show two peaks, similar to that of OM. FID chromatogram profile confirmed that both peaks contain peptide. The high molecular conjugate was probably formed from OM aggregate by cross-linking of OM molecules with Lys-Lys peptide moiety. The peptide content in conjugates was estimated by UV spectroscopy to be (28 ± 5) mass % or (3.0 ± 0.5) mole %. Free peptide in the product was below the detection limit e.g. less than 0.2 mass %.

Preparation of labelled mannan–peptide conjugate

The aim was to prepare labelled mannan and mannan–peptide conjugate with the same ratio of components. For this reason, the labelled mannan–peptide conjugate was prepared from a portion of mannan–peptide conjugate reaction mixture before quenching by sodium borohydride. The pH was adjusted to 4.5 with acetate buffer and required amount of 2AB was added. Otherwise, the reaction proceeds similarly to 2AB-mannan preparation.

The conjugate was analysed by HPSEC with RI (Fig. 1b) and FI detector (Figs. 2a and 2b). Conjugate and cross-linked conjugate was observed on the chromatograms however no free peptide was observed (Fig. 2b). The peptide and 2AB content in the conjugates were estimated by UV spectroscopy to be (27 ± 3) mass % or (2.9 ± 0.3) molar % and (2.2 ± 0.1) mass % or (2.7 ± 0.1) molar %, respectively. Free peptide in the product was below the detection limit.

The prepared labelled mannan and mannan–peptide conjugates were in sufficient yields (see scheme in Fig. 3). By precise control of the reactions conditions, it was possible to minimize the polysaccharide chain degradation. The peptide and 2AB fluorescent label contents were in very good agreement with conjugate design reaction settings. In terms of incorporation efficiency (substrate bound/substrate added), it was ≥93 % for 2AB and ≥89 % for peptide. The cross-linked byproducts were not isolated and characterised. This task can be solved in the future.

Fig. 3
figure 3

Reaction scheme of the conjugates preparation. The oxidation of terminal mannose C6 hydroxyl groups to the aldehyde groups is followed by reductive amination of the peptide or 2-aminobenzamide. The reaction sequence is quenched by complete reduction of remaining aldehyde groups back to hydroxyls

Full size image

Fluorescence cell imaging

The time-dependent accumulation of mannan–2AB and mannan–peptide–2AB in intracellular organelles of RAW 264.7 cells was assessed by fluorescence cell imaging. The fluorescence intensity of mannan–2AB and mannan–peptide–AB deposits in organelles following the mannan–2AB and mannan–peptide–2AB exposition reflected the cell transport processes (Fig. 4).

Fig. 4
figure 4

Fluorescence pattern of mannan–2AB (a, b) and mannan–peptide–2AB (c, d) uptake by RAW 264.7 cells. In situ fluorescence has been evaluated using Imager A.1, magnification 630×, AxioVision, Zeiss, Germany

Full size image

Immunocytometry

Mannan is specifically recognised by pattern-recognition receptors (PRR) as MMR, TLR-4, TLR-2, TLR-6, Dectin-2, Galectin3, DC-Sign [25, 26], the recognition of mannan–peptide could be realised by different mechanism as opposed to mannan alone [2] and could initiate different signalling pathways. There are several pattern-recognition receptors for mannan on RAW 264.7 cells e.g. MMR, TLR-4 [27]. Concerning, that the experiments were performed with not natural high-molecular-mass mannan, but with ultrasonicated and chemically modified mannan, it could be only assumption that pathogen-associated molecular patterns–PRR interactions will reflect the interactions of native untreated mannan. The elucidation of labelled mannan and mannan–peptide conjugate internalization and intracellular processing requires further investigation. Nevertheless, the 2AB labelled mannan or mannan–peptide conjugate represent the useful basis for investigation of conjugate processing.

Taking into account these facts we performed another experiment with mannan labelled with FITC, with the possibility of quantitative measurement of the extracellular (surface bound) and intracellular fluorescence using immunocytometry and trypan blue specific quenching for FITC which is not possible for 2-AB. As could be seen from Fig. S4, based on mean fluorescence intensity without quenching and after quenching of FITC fluorescence, it is possible to analyse the portion of labelled mannan accumulated intracellularly parallel to examination of extracellular portion. Moreover this inclusion and transport can be followed in time and concentration dependent manner.

The time and concentration dependent mannan–FITC cell-processing (Fig. S4) revealed sequential loss of extracellular (cell surface) mean fluorescence intensity with parallel intracellular accumulation. These processes are highly concentration dependent. Highest concentration of 0.25 mg mannan–FITC/105cells seems to be the most effective. This is supported by visualising via fluorescence microscopy (Fig. S5).

Conclusions

We successfully prepared well defined low-molecular-mass mannan with narrow molecular mass distribution containing selectively oxidized C6 to aldehyde. The proposed method of labelled mannan and mannan–peptide conjugates preparation is relatively simple and fast. The prepared conjugates were well characterized and contained no detectable unreacted protein. The biggest advantage of the method is the possibility to obtain a product with the desired composition by simple adjustment of the reaction mixture conditions. In addition, we demonstrated by in vitro fluorescence visualization of the 2AB labelled mannan or mannan–peptide conjugate the accumulation of prepared constructs in intracellular compartment of RAW 264.7 cells.

We applied the 2AB for the labelling of mannan as carbohydrate antigen and mannan-peptide conjugate. The 2AB was mainly used to label carbohydrates in HPLC experiments. These newly prepared products can serve as appropriate model for cell tracking investigation. Both the fluorescence and UV absorption are high and they are desired for precise characterisation of conjugate and intracellular tracking. In summary, the procedure is far more selective than classical ones (e.g. periodic acid vs. Dess–Martin oxidation), the amount of bound fluorescent label (2AB) can be precisely defined through the amount that is added into the reaction.