GM1 Oligosaccharide Crosses the Human Blood-Brain Barrier In Vitro by a Paracellular Route.

Ganglioside GM1 (GM1) has been reported to functionally recover degenerated nervous system in vitro and in vivo, but the possibility to translate GM1's potential in clinical settings is counteracted by its low ability to overcome the blood-brain barrier (BBB) due to its amphiphilic nature. Interestingly, the soluble and hydrophilic GM1-oligosaccharide (OligoGM1) is able to punctually replace GM1 neurotrophic functions alone, both in vitro and in vivo. In order to take advantage of OligoGM1 properties, which overcome GM1's pharmacological limitations, here we characterize the OligoGM1 brain transport by using a human in vitro BBB model. OligoGM1 showed a 20-fold higher crossing rate than GM1 and time-concentration-dependent transport. Additionally, OligoGM1 crossed the barrier at 4 °C and in inverse transport experiments, allowing consideration of the passive paracellular route. This was confirmed by the exclusion of a direct interaction with the active ATP-binding cassette (ABC) transporters using the "pump out" system. Finally, after barrier crossing, OligoGM1 remained intact and able to induce Neuro2a cell neuritogenesis by activating the TrkA pathway. Importantly, these in vitro data demonstrated that OligoGM1, lacking the hydrophobic ceramide, can advantageously cross the BBB in comparison with GM1, while maintaining its neuroproperties. This study has improved the knowledge about OligoGM1's pharmacological potential, offering a tangible therapeutic strategy.


N2a cells
N2a cells are not listed as a commonly misidentified cell line by the International Cell Line Authentication Committee. N2a cells were bought from Sigma-Aldrich to which they were supplied by European Collection of Authenticated Cell Cultures (ECACC) (Catalogue No. 89121404; Lot No. 13K010, passage +9). N2a were used from passage +10 to passage +15 to conduct experiments reported in the present manuscript.
To verify the authentication of employed N2a cells we performed the following tests at the beginning and end of single experimental work.


Morphology check by microscope To identify the state of cells, we checked cellular morphology by phase contrast microscopy (Olympus BX50 microscope; Olympus, Tokyo, Japan). Morphological outcomes of N2a cells confirmed the expected neuronal/ameboid-like morphology (data not shown).  Growth curve analysis Cell proliferation was evaluated according to MTT method [1,2]. Briefly, 2.4 mM MTT (4 mg/ml in PBS) were added to each well and plates were re-incubated for 4 h at 37°C. Medium was carefully removed and replaced with 2-propanol: formic acid, 95:5 (v/v). Plates were gently agitated prior to read the absorbance at 570 nm with a microplate spectrophotometer (Wallac 1420 VICTOR2TM, Perkin Elmer). The growth profile showed a normal growth rate (data not shown).  Mycoplasma detection Mycoplasma infection was evaluated by fluorescent Hoechst staining [3], a fluorescent dye that binds specifically to DNA and that reveals the presence of mycoplasma infections as intracellular particulate or filamentous fluorescence at 400X magnification using NikonEclipse Ni upright microscope. The mycoplasma test has always given negative results (data not shown).

Caco-2 cells
Caco-2 cells are not listed as commonly misidentified cell line by the International Cell Line Authentication Committee. However it is reported that culture condition and cell origin can deeply affect Caco-2 morphology and behavior [4][5][6]. Thus, we deeply verified and authenticated our Caco-2 cultures. We used Caco-2 cells during only ten passage (from passage +26 to +36) as previously reported [4].  Morphology check by microscope We verified that our Caco-2 cells form compact and homogenous monolayer of mostly small diameter cells (data not shown).  Mycoplasma detection Since mycoplasma could infect of Caco-2 cells, we validated the absence of any contamination by two methods (data not shown). As for N2a cells we directly staining nuclei of Caco-2 cells by fluorescent Hoechst staining. Moreover, we confirmed the absence of mycoplasma, by using the specific Lonza mycoplasma assay. Briefly culture supernatants were mixed with MycoAlert reagents according to manufacturer instruction and the luminescence was measured after 10 min using fluorimeter (BioTek, H1, Vermont, Winooski).

Morphological analysis for neurite outgrowth evaluation
N2a cells, treated or not with hBLEC-crossed OligoGM1 were observed by phase contrast microscopy (200X magnification, Olympus BX50 microscope; Olympus, Tokyo, Japan). At least 10 fields from each well were photographed for each experiment.

Evaluation of OligoGM1 toxicity on Caco-2 cells
Viability of Caco-2 cells after OligoGM1 treatments was determined by MTT assay (as previously reported [1,7]. Briefly, after 1 h from OligoGM1 incubation, cells plated in a 96-well were washed and incubated with 100 μL of 2.4 mM MTT (4 mg/mL in RH) for 1 h at 37°C in a humidify atmosphere of 95% air / 5% CO2. Subsequently MTT was carefully removed and replaced with DMSO convert MTT into violet formazan. The quantity of formazan was measured by measuring the absorbance at 570 nm using a plate reading spectrophotometer (BioTek, H1, Vermont, Winooski) which was compared with a control condition (untreated Caco-2 cells) after cell solubilization with DMSO.

Apparent permeability (Papp)
Papp index is widely used to screen the absorption process of drugs [4,[8][9][10]. It is calculated according to the equation (1): where (∆Q/∆t) (μmol/sec) is the rate of permeation (quantity per second) of a molecule across the cell layer, C0 (μmol/cm 3 ) is the donor compartment concentration at time zero and S (cm 2 ) is the area of the cell monolayer.
In the transwell insert system there are two compartments and, analyzing the absorption of a substance, the apparent permeability index can be obtained for the apical to basolateral transport, Papp A  B, and for the inverse, basolateral to apical transport Papp B  A.
To define Papp (1), the Fick's law could be used [11], which describes the transport of substances through a semipermeable membrane. Given that the initial concentration of the solute in the donor compartment is much higher than the one in the receiver compartment (equal to 0 at t0), and assuming that the solute equally partitions in the two compartments, for small variation of C0, it is possible to integrate and simplify the Fick's equation to obtain as follows: where ∆Q/∆t corresponds to the amount of solute that crosses the membrane over time, D is the diffusion coefficient of the substance in water, S is the membrane surface, h its thickness and C0 is the concentration of the solute which approximates the initial concentration C0.
The equation (2) can be rewritten as: equivalent to Papp (1), which can thus be defined as the ratio between D and the thickness h of the membrane. D is defined by the Stokes-Einstein [12] as follows: where kB is the Boltzmann constant, T is the absolute temperature and η the medium viscosity. This equation is particularly interesting because it relates the molecule mobility to its hydrodynamic radius (r). Thus according to equations (1)(2)(3)(4), the Papp of a solute directly depends on its hydrodynamic radius (r). Assuming that two different molecules at the same concentration move passively from one compartment to another, the relationship between their Papp at a given temperature defined in (4) will be equal to the inverse of the respective hydrodynamic radius values, as follows: Using the equation above we can now directly compare the permeability of different molecules as long as their hydrodynamic radius are known.

Endothelial permeability (Pe)
Pe is calculated to appreciate more accurately the transport of a drug across an endothelial barrier. In fact, this approach, based on the clearance principle, allows to obtain a result independent from the administered concentration [23,24]. In order to obtain the increment in cleared volume between successive sampling events, the amount of solute transport during a time interval is divided by the donor chamber concentration. The total cleared volume is the sum of the incremental cleared volume up to the considered time points, according to the equation (6): where Q (pmol) is the amount of a drug in the receiver compartment and Cd is the donor chamber concentration at each time point (pmol / mm 3 ). If during the experiment time, the clearance volume increases linearly, the average volume cleared can be plotted versus time. The slope, resulted by linear regression analysis of clearance, C (mm 3 ), and time, t (min), represents the permeability associated to a surface of transport: The value can be obtained from endothelial cell-covered filters, PSt and from inserts covered only by Matrigel, PSf. Once elaborated this value, the endothelial permeability can be derived applying the equation (8): and finally the following formula (9), that involves the measure of the transport surface, S (cm 2 ): = (10) Table S1   Table S1. Direct transport related endothelial permeability (Pe A  B) (×10 -3 cm/min) of LY (50 μM) alone or in combination with GM1 and OligoGM1 at different concentration (10, 50, 100 and 300 μM) after 60 min. Results are mean ± SEM from 4 independent experiments (n = 4) examining a minimum of 3 well for each condition.    RH buffer Transfectragro medium