Construction of Nicotinamide Mononucleotide-Loaded Liposomes and Their In Vitro Transport Across the Blood–Brain Barrier

Abstract

Nicotinamide mononucleotide (NMN) possesses a variety of physiological functions and has therapeutic effects on cardio-cerebral diseases, senile degenerative diseases, neurodegenerative diseases, delayed aging, etc. However, its ability to cross the blood–brain barrier (BBB) and the mechanism of its transport have not been reported. In this study, we used the immortalized hCMEC/D3 cell line to construct an in vitro monolayer cell BBB model, evaluated its ability to cross the blood–brain barrier, and explored the mechanism by carrying out transport and efflux experiments on NMN. The ability of NMN to cross the BBB was investigated by preparing NMN-loaded liposomes conjugated with ANG peptide and RVG peptide. The results showed that the transmembrane transport ability of NMN was moderate, and the transport mechanism was passive transport relying on the concentration difference. The trans-BBB ability of ANG peptide coupled with NMN could be highly significantly improved.

1. Introduction

Nicotinamide mononucleotide (NMN), a naturally occurring biologically active nucleotide, can be naturally synthesized in cells and is also found in a variety of plants and animals [1,2,3], such as hairy beans, broccoli, shrimp, etc., with hairy beans containing the highest amount of NMN, up to 1.88 mg/100 g [4]. Studies have shown that NMN has a variety of physiological activities and has therapeutic effects on brain diseases such as Alzheimer’s disease, Parkinson’s disease, and cerebral hemorrhage [5,6,7]. The prerequisite for a drug to act on the brain and have therapeutic effects is that the drug reaches the site of action. As a highly specialized system of endothelial cells (mainly including brain endothelial cells, astrocytes, pericytes, tight junctions, etc.) that exists between circulating blood and central extracellular fluid, the blood–brain barrier (BBB) is a key barrier separating the blood from the underlying brain cells and is limiting for the delivery of drugs for the treatment of brain diseases [8,9,10,11]. However, to our best knowledge, the ability of NMN to cross the blood–brain barrier (BBB) and the mechanism remains unclear.

To date, the immortalized hCMEC/D3 cell line has become a useful model for screening drug candidates in the central nervous system [12,13]. This cell line has the major advantages of a high proliferation rate and phenotypic invariance during passaging and retains most of the morphological and functional characteristics of brain endothelial cells even when cultured alone [14,15], which is more suitable for drug uptake and pathological modeling studies. The main pathways of drugs across the blood–brain barrier are passive transport, active transport, and cytosolization and cytotranspiration [16]. Although some drugs can penetrate the blood–brain barrier, drug-metabolizing enzymes and drug transporters on the blood–brain barrier inhibit the diffusion of the drug to the therapeutic site, resulting in poor bioavailability of the drug in the central nervous system [17,18]. Research indicates that liposomes modified with particular peptides have demonstrated remarkable effectiveness in promoting the passage of molecules across the blood–brain barrier [19,20,21]. This strategy shows promise for delivering drugs to the central nervous system, and our study intends to advance this field of NMN-focused research.

In this study, the immortalized hCMEC/D3 cell line was selected to construct an in vitro cell monolayer BBB model to investigate the transmembrane capacity of NMN across the BBB and its transport mechanism. Moreover, we constructed an NMN-loaded liposome and modified the liposome surface with two short peptides, ANG and RVG, to enhance blood–brain barrier penetration and target nanovesicles to gliomas [22,23,24]. ANG peptides have a high affinity for low density lipoprotein receptor-related protein (LRP), which is overexpressed in both the blood–brain barrier and gliomas [22], and studies have shown that ANG peptides can enhance blood–brain barrier penetration and target nanovesicles to gliomas [22,23,24]. By evaluating their ability to penetrate the BBB, this study could lay a basis for the study of NMN in the development of new drugs in the central nervous system and provide new ideas for research related to NMN-targeted delivery.

2. Materials and Methods

2.1. Materials

Nicotinamide Mononucleotide (95%), sodium fluorescein, DSPE-PEG2000(1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000]), and DSPE-PEG2000-Mal (1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[maleimide(polyethylene glycol)-2000]) were obtained from Shanghai yuanye Bio-Technology Co., Ltd. (Shanghai, China). Soy lecithin was obtained from Avanti (Alabaster, AL, USA). The peptide RVG with cysteine at the C-terminus (H2N-YTIWMPENPRPGTPCDIFTNSRGKRASNG-OH, HS-RVG) and the peptide ANG with cysteine at the C-terminus (TFFYGGSRGKRNNFKTEEYC, HS-ANG) were synthesized by MeloPEG (Shenzhen, China). β-Mercaptoethanol (β-ME), cholesterol was obtained from Sigma-Aldrich (St. Louis, MO, USA), and 3-(4,5-Dimethyl-2-Thiazolyl)-2,5-Diphenyl Tetrazolium Bromide (MTT) was obtained from Thermo Fisher Scientific (Waltham, MA, USA). Other chemicals were of analytical grade, and solvents used in HPLC were of HPLC grade.

Transwell^TM plates of 12 wells (12 mm, pore size 3.0 μm) were purchased from Corning Costar (Cambridge, MA, USA). Endothelial Cell Medium was obtained from ScienCell Research Laboratories (Carslbad, CA, USA). Hanks’ balanced salts solution (HBSS) was obtained from Gibco Laboratories (Life Technologies Inc., Grand Island, NY, USA).

hCMEC/D3 cell line was obtained from Hunan Fenghui Biotechnology Company (Changsha, China).

2.2. Liposome Preparation

The liposomes were prepared through the thin-layer-film method. Briefly, soy lecithin (100.00 mg), DSPE-PEG2000 (13.32 mg), DSPE-PEG2000-Mal (8.48 mg), and cholesterol (14.00 mg) were dissolved in 10 mL of dichloromethane/methanol (2:1, v/v) solution to a 50 mL round bottom flask [25]. Then, the solvent was completely evaporated under reduced pressure to form a thin film. The thin film was completely hydrated with 5 mL of NMN PBS buffer (NMN concentration of 20 mmol/L) in the dark at 37 °C for around 60 min to obtain a lipid suspension. The lipid suspension was ultrasonicated for 5 min at 300 W, followed by at least 10 rounds of extrusion through a polycarbonate filter membrane (pore size, 100 nm) with the liposome extruder at 50 °C. The unloaded NMN was removed by ultrafiltration (10 KDa cutoff), and the concentrated lipid solution was obtained (Figure 1).

HS-RVG (1 mg/mL) and HS-ANG (1 mg/mL) were firstly reduced by βME (5 equal molar amounts) in aqueous solution for 4 h at room temperature and then reacted with the Mal groups on the surface of the prepared liposome at the 1 to 1 molar ratio of total sulfur [HS-RVG/ANG and β-mercaptoethanol] to Mal [24]. After the reaction, the β-ME and unreacted material were removed by ultrafiltration (10 KDa cutoff), and the concentrated-targeted liposome (ANG-NMN-lips and RVG-NMN-lips) stock solutions were obtained (Figure 1).

2.3. Measurement of Liposome Particle Size

The particle size and size distribution were measured by Malvern Zetasizer Nano ZS dynamic light scattering (DLS). The liposome suspension was diluted with citric acid sodium dihydrogen phosphate buffer solution to the concentration of liposomes to 0.1 w/v%.

2.4. Measurement of Encapsulation Rate and Drug Loading

The liposome suspension (0.5 mL) was mixed with methanol (2 mL) and incubated for 5 min at 60 °C to destroy the liposome structure and promote the dissociation of NMN encapsulated in the liposome. The content of the released NMN (M_in) was measured with HPLC. At the same time, the unencapsulated NMN (M_out) from the filtrate of ultrafiltration was measured directly with HPLC. The solid mass of the liposome suspension (M) was obtained by freeze-drying.

The encapsulation rate and the drug loading of liposomes is calculated as follows [26]:

$$\mathrm{Encapsulation}\mathrm{rate}(\%)={\displaystyle \frac{{\mathrm{M}}_{\mathrm{i}\mathrm{n}}}{{\mathrm{M}}_{\mathrm{i}\mathrm{n}}+{\mathrm{M}}_{\mathrm{o}\mathrm{u}\mathrm{t}}}}\times 100\%$$

(1)

$$\mathrm{Drug}\mathrm{loading}(\%)={\displaystyle \frac{{\mathrm{M}}_{\mathrm{i}\mathrm{n}}}{\mathrm{M}}}\times 100\%$$

(2)

2.5. Preparation of hCMEC/D3 Monolayer

The hCMEC/D3 cell line was cultured in ECM complete medium containing D-glucose (4.5 g/L), NaHCO₃ (3.7 g/L), supplemented with 10% FCS, 1%NEAA, penicillin (100 U/mL), and streptomycin (100 μg/mL) in an atmosphere of 5% CO₂ and 95% relative humidity at 37 °C. All cells used in this study were between 25 and 35 passages with a variation of ten passages [27].

The hCMEC/D3 cells were seeded at a density of about 7 × 10⁴ cells/cm² on a 12-well Transwell^TM insert filter and left to grow for 7 days to reach confluence and differentiation. The integrity of the cell monolayer was monitored by transepithelial electrical resistance (TEER) with an epithelial voltammeter (World Precision Instrument, Sarasota, FL, USA). Transport control assays were run using sodium fluorescein as paracellular [28].

2.6. Transport Experiments

2.6.1. Cytotoxicity Assay

The MTT assay was used to examine the toxicity of NMN, blank liposomes (KB lips), drug-loaded liposomes (NMN-lips), ANG-NMN-lips, and RVG-NMN-lips on hCMEC/D3 cell lines. The hCMEC/D3 cells were seeded in 96 well plates at a density of 1 × 10⁴ cells per well with 200 mL of culture medium for 24 h. The stock solutions of NMN, KB-lips, NMN-lips, and ANG-NMN-lips were diluted with ECM complete culture medium to the testing concentrations of 0.02 mmol/L, 0.05 mmol/L, 0.1 mmol/L, 0.2 mmol/L, 0.5 mmol/L, 1 mmol/L, 2 mmol/L, and 5 mmol/L, respectively. And these testing media were added into the wells (with 4 replicates for each concentration). No treated cultures served as controls. After 24 h treatment, the medium was removed, and 0.1 mL of MTT solution (5 mg/mL) was added to each well to react for 45 min at 37 °C. The supernatant was removed, and 0.1 mL DMSO was added to dissolve the MTT formazan crystals. The plates were gently shaken for 10 min until the complete solvation of the purple formazan crystals. Absorbance at 570 nm was measured and the results were normalized by controls.

Cell viability (%) = (OD_drug/OD_{average value of blank group}) × 100%

(3)

2.6.2. Transport and Efflux Experiments

The cell monolayer was used for the transport experiments when it had been cultured for 7 days and the TEER value was 30~40 Ω·cm².

After gently washing the hCMEC/D3 cell monolayer twice with prewarmed HBSS medium, the test substance solutions were added to either the apical (AP, 0.5 mL) or the basolateral (BL, 1.5 mL) side of the cell monolayer. After incubating the cell monolayer in a cell culture incubator at 37 °C and 5% CO₂ for a certain time, samples (400 μL on the AP side or 1400 μL on the BL side) were collected, immediately frozen, lyophilized, and preserved below −80 °C before analysis.

After processing, the optimized HPLC method was used for determination, and the P_app value was calculated.

P_app (cm/s) = V_B/A·C_A0 × (△C_B/△t)

(4)

where P_app is the apparent permeability coefficient (cm/s); V_B is the base side volume; A is the surface area of the insert (cm²); C_A0 is the initial concentration of the test compound on the donor side (mmol/mL); and △C_B/△t refers to changes in concentration on the outer side of the substrate over time. All experiments were carried out in triplicate. Data are the mean ± SD.

2.7. Intracellular Accumulation and Recovery

For the quantification of the test substance accumulated inside the hCMEC/D3 cells, the lyophilized cells were sonicated with 200 μL 50% methanol for 20 min and then centrifuged at 15,000× g for 10 min. An aliquot of 10 μL of the supernatant was used for the assay.

The recovery of the test NMN was measured at both sides of the insert and intracellular accumulation of the hCMEC/D3 cells.

$$\mathrm{Recovery}(\mathrm{\%})={\displaystyle \frac{{\mathrm{C}}_{\mathrm{A}\mathrm{f}}{\mathrm{V}}_{\mathrm{A}}+{\mathrm{C}}_{\mathrm{B}\mathrm{f}}{\mathrm{V}}_{\mathrm{B}}}{{\mathrm{C}}_{\mathrm{A}0}{\mathrm{V}}_{\mathrm{A}}}}\times 100\%$$

(5)

where C_Af is the AP side final concentration; C_Bf is the BL side final concentration; V_A is the AP side volume; V_B is the BL side volume; C_A0 is the initial concentration on the AP side.

2.8. Analytical Method

2.8.1. Sample Pre-Treatment

After freeze-drying the samples, they were dissolved in 100 μL of 50% methanol-water, vortexed and mixed well, and centrifuged at 15,000× g for 10 min. An aliquot of 10 μL of the supernatant was used for the assay.

2.8.2. Establishment of HPLC Conditions

Based on a developed method [5], the chromatographic conditions were optimized, and the effects of the mobile phase ratio, flow rate, column temperature, and injection volume on the chromatographic peaks in HPLC were explored to determine the HPLC conditions.

The stability of HPLC was evaluated by studying the linear relationships, precision, repeatability, stability, and sample recovery rate.

2.9. Data Processing

All experiments were repeated three times. GraphPad Prism 9.0.0 software and Origin 2018 64Bit software were used to organize, analyze, and display the data involved in the above experiment. The measurement data were expressed in the form of mean ± standard deviation (͞x ± SD), and t-test analysis was performed on the data. p < 0.05 was considered significant, and p < 0.01 was considered extremely significant.

3. Results

3.1. Preparation and Characterization of NMN-Lips

The obtained liposome solution appears as a slightly hazy transparent suspension (Figure 2a). The particle sizes of NMN-lips and KB-lips were measured to be 100.10 ± 40.72 nm and 127.75 ± 57.26 nm. And the particle sizes of ANG-NMN-lips and RVG-NMN-lips were 132.50 ± 85.86 nm and 139.28 ± 84.83 nm, respectively, both of which are larger than NMN-lips, indicating the successful connection of the two peptides (Figure 2b).

The PDI values of KB-lips, NMN-lips, ANG-NMN-lips, and RVG-NMN-lips are 0.1226, 0.2027, 0.1596, and 0.1280, respectively, indicating that the distribution of the four liposomes is relatively uniform, and the initial particle sizes and polydispersity were adequate for the study’s objective.

3.2. Optimization of HPLC Method for Detecting NMN

A rapid, simple, and reliable HPLC method has been established for the analysis of the NMN in the hCMEC/D3 cell cultures by adjusting the HPLC mobile phase to a 0.1% formic acid aqueous solution—a 0.1% formic acid methanol solution (v/v = 5:95) (

Table 1. Optimized chromatographic conditions.

Chromatographic Conditions	Result
Pump	Binary pump
Detector	Ultraviolet detector
Chromatographic Column	Venusil HILIC (4.6 × 250 mm, 5 μm)
Mobile Phase A	0.1% formic acid aqueous solution
Mobile Phase B	0.1% formic acid methanol solution
A:B	5:95 (v/v), isocratic elution
Detection Wavelength	235 nm
Current Speed	1 mL/min
Column Temperature	35 °C
Injection Volume	10 μL

). Under the optimized conditions, good linear regression, good instrument precision, good experimental reproducibility, and good sample recovery rate can be obtained. The standard calibration curves were constructed by plotting the peak area (y) versus the concentration (x, mM). The linear equation and range were y = 3569.9x − 154.25 (r² = 0.9998) and 0.05–5.00 mmol/L for 1. The RSD values of the precision, repeatability, stability, and sample recovery rate are all less than 2%, making it an effective method for determining NMN content in actual biological samples.

3.3. Evaluation of In Vitro Monolayer BBB Model

The TEER value can be used to evaluate the integrity of the cell monolayer (Figure 3). Within the first 6 days of culturing cell lines in polycarbonate membrane chambers, the TEER value of the cell monolayer gradually increased with time, and the increasing trend was approximately linear. On the 7th day, the TEER value was 35.28 Ω·cm², and on the 8th day, the TEER value was similar to that on the 7th day, at 35.21 Ω·cm², indicating that the TEER value plateau period began on the 7th day (Figure 3). The TEER values during the plateau period are similar to those reported in reference [29], providing preliminary evidence that a single layer of blood–brain barrier cells can be used for subsequent experiments.

On the 7th day of cell line culture in a polycarbonate membrane chamber (TEER = 36.06 Ω·cm²), fluorescein sodium permeability experiments were conducted to evaluate the integrity of the cell monolayer. The transport ability of sodium fluorescein and the P_app value was measured by the method discussed in Section 2.6.2.

Table 2. The measured values of sodium fluorescein (2 h) and the P_app values reported in the literature (n = 3).

Group	Papp Value (cm/s) ($\overline{\mathbf{x}}$ ± SD)
Actual measurement value	29.2 × 10−6 ± 0.65
Literature report value [28]	25.8 × 10−6 ± 1.66

shows the comparison results between the measured values of fluorescein sodium at 2 h of administration and the reported P_app values of fluorescein sodium in the literature. It can be seen that the measured values are similar to the reported values in the literature, i.e., of the same order of magnitude, further proving the success of in vitro monolayer BBB modeling.

This model can be used for subsequent transmembrane transport experiments to study the transmembrane ability and transport mechanism of NMN- and NMN-lips.

3.4. Results of Cytotoxicity Assay

The MTT assay was used to investigate the cytotoxicity of NMN on hCMEC/D3 cells. When the concentration of NMN was lower than 2 mmol/L, the viability of hCMEC/D3 cells was greater than 90% (Figure 4a), indicating the low toxicity of NMN. We also investigated the toxicity of different types and concentrations of liposomes on hCMEC/D3 cells. As we can see from Figure 4b, when the concentration of the blank liposome was lower than 2 mmol/L, the viability of hCMEC/D3 cells was greater than 90%. And the viabilities of hCMEC/D3 cells were greater than 90%, when the administration concentration of NMN-lips (Figure 4c), ANG-NMN-lips (Figure 4d), and RVG-NMN-lips (Figure 4e) were lower than 0.2 mmol/L, 0.5 mmol/L, and 0.5 mmol/L, respectively. As such, the drug concentration in subsequent uptake and bidirectional transmembrane transport experiments was controlled to be lower than 0.2 mmol/L.

3.5. Research on the Transmembrane Transport Mechanism of NMN

The bilateral permeabilities of the NMN were investigated. The concentration of NMN was 0.2 mmol/L. The incubation time was up to 120 min. The apparent permeability coefficient (P_app) and recovery rate of NMN were calculated (

Table 3. Results of NMN transmembrane transport (n = 3).

Sample	Papp  (10−6, cm/s)  ( $\overline{\mathbf{x}}$ ± SD)		AP-BL/ BL-AP	BL-AP/ AP-BL	Recovery (%)
	AP-BL	BL-AP			AP-BL	BL-AP
NMN	7.31 ± 0.31	7.40 ± 0.13	0.988	1.012	84.34 ± 2.33	93.41 ± 1.62

). The P_app value of the AP-BL transport experiment in the NMN bidirectional transmembrane is (7.31 ± 0.31) × 10⁻⁶ cm/s, and the P_app value of the BL-AP efflux experiment is (7.40 ± 0.13) × 10⁻⁶ cm/s. The P_app values of bidirectional transmembrane transport are both on the order of 10⁻⁶, indicating that the transport capacity of NMN is moderate. The recovery rate of the AP-BL transport experiment was 84.34 ± 2.33%, and the recovery rate of the BL-AP efflux experiment was 93.41 ± 1.62%, both within the range of 80% to 120%, indicating no metabolism. Furthermore, we calculated the ratio of P_app values in NMN transport and efflux experiments. The P_app value ratio of AP-BL/BL-AP was 0.988; the P_app value ratio of BL-AP/AP-BL is 1.012. The ratio of P_app values < 1.5 indicates that NMN is not affected by efflux in the hCMEC/D3 cell model, and the ratio of P_app values is approximately equal to 1. It is speculated that the transmembrane mechanism of NMN crossing the BBB of monolayer cells in vitro is passive transport, and NMN is transported across the membrane through concentration differences. No active transport was found.

The time- and concentration-dependency bilateral permeabilities of the NMN were investigated. To observe the time dependency, 0.2 mmol/L of NMN was added to the donor side and was sampled from 0 to 180 min. To observe the concentration dependency, the bidirectional transport rates of NMN were measured in the concentration range of 0.05–2.00 mmol/L, and the time was set as 120 min. To check the mass balance, the recoveries of the NMN were measured at both sides of the insert and their intracellular accumulation within the hCMEC/D3 cells.

The transport kinetics curve and transport percentage in the NMN time-dependency bidirectional transmembrane transport experiment are shown in Figure 5. It can be seen that, when the administration concentration is 0.2 mmol/L, the amount of NMN transported substances at the receiving end shows an increasing trend followed by a stable trend with time. This is because, as time goes on, the transport of NMN reaches saturation, and the increase in the amount of NMN transported substances is relatively small from 90 min to 180 min, and the maximum amount of transported substances is reached at 120 min, which is 56.19 × 10⁻⁹ mol. Therefore, 120 min is the optimal time for all subsequent experiments. The transport percentage result is similar to its kinetic curve and tends to stabilize with the increase in time. The growth rate of NMN transport percentage is relatively small, from 90 min to 180 min, because it has reached saturation state.

Table 4. Recovery rate results of NMN time-dependency bidirectional transmembrane transport experiment (n = 3).

Sample	Time	Recovery (%) ($\overline{\mathbf{x}}$ ± SD)
		AP-BL	BL-AP
NMN	30 min	97.94 ± 1.38	92.81 ± 1.27
	60 min	99.86 ± 1.73	94.74 ± 1.46
	90 min	102.7 ± 2.45	96.52 ± 3.07
	120 min	103.6 ± 1.84	86.63 ± 1.47
	150 min	97.91 ± 2.20	87.76 ± 1.76
	180 min	89.12 ± 0.51	86.71 ± 1.40

shows the recovery rate results in the NMN time-dependency bidirectional transmembrane transport experiments. It can be seen that the recovery rate of NMN in the in vitro BBB model constructed in this study was within the range of 80–120% at six time points from 30 to 180 min, indicating that NMN was not metabolized in the model constructed in this study.

The transport kinetics curve and transport percentage in the NMN time-dependency bidirectional transmembrane transport experiment are shown in Figure 6. It can be seen that, when the administration time is 120 min, the amount of substance transported by NMN at the receiving and administering ends increases linearly with increasing concentration, and the higher the concentration, the higher the amount of substance transported. The amount of substance transported between the receiving end and the administration end increases linearly with the increase in the NMN concentration, and the higher the concentration, the greater the amount of substance transported. The percentage of NMN transport showed a decreasing trend and tended to stabilize with increasing concentration. When the concentration of administration is greater than 0.5 mmol/L, there is almost no difference in the transport rate of NMN, indicating that saturation has been reached. Taking into account the experimental results of AP-BL and BL-AP in the concentration-dependency bidirectional transmembrane transport experiment, it is determined that the drug concentration for subsequent experiments should be less than 0.5 mmol/L.

Table 5. Recovery results of NMN concentration-response bidirectional transmembrane transport experiment (n = 3).

Sample	Concentration (mmol/L)	Recovery (%) ($\overline{\mathbf{x}}$ ± SD)
		AP-BL	BL-AP
NMN	0.05	114.4 ± 1.10	109.8 ± 0.86
	0.1	110.0 ± 0.93	101.0 ± 0.29
	0.2	104.3 ± 1.70	101.8 ± 3.42
	0.5	86.34 ± 0.66	93.01 ± 1.06
	1	84.43 ± 0.96	87.45 ± 0.48
	2	83.80 ± 0.50	93.01 ± 0.63

shows the recovery results of the NMN concentration-dependency bidirectional transmembrane transport experiment. It can be seen that, in the six concentration nodes from 0.05 mmol/L to 2 mmol/L, the recovery rate of NMN in the model constructed in this study was within the range of 80% to 120%, indicating that NMN was not metabolized in the model constructed in this study.

3.6. Liposome Bidirectional Transmembrane Transport Experiment

The bilateral permeabilities of the NMN, NMN-lips, RVG-NMN-lips, and ANG-NMN-lips were investigated. At the same concentration, the P_app values of NMN, NMN-lips, and RVG-NMN-lips are on the order of 10⁻⁶, indicating that their transport ability on the in vitro monolayer BBB constructed in this study is moderate; the P_app values of ANG-NMN-lips is on the order of 10⁻⁵, indicating good transport capacity (

Table 6. Results of NMN transmembrane transport (n = 3).

Sample	Papp Value (10−6, cm·s−1) ($\overline{\mathbf{x}}$ ± SD)		Ratio of AP-BL/BL-AP	Ratio of BL-AP/AP-BL
	AP-BL	BL-AP
NMN	7.31 ± 0.31	7.40 ± 0.13	0.988	1.012
NMN-lips	7.44 ± 0.70	7.26 ± 0.29	1.024	0.977
RVG-NMN-lips	8.88 ± 0.92	7.67 ± 0.10	1.129	0.886
ANG-NMN-lips	10.71 ± 0.77	9.49 ± 0.13	1.157	0.864

). We calculated the ratio of P_app values based on the transmembrane transport of AP-BL and BL-AP by NMN, NMN-lips, RVG-NMN-lips, and ANG-NMN-lips (Table 6). Within 2 h of administration, the P_app values of AP-BL/BL-AP for NMN and NMN-lips, RVG-NMN-lips, and ANG-NMN-lips were 0.987, 1.024, 1.129, and 1.157 respectively; the P_app value ratios of BL-AP/AP-BL are 1.013, 0.977, 0.886, and 0.864 respectively. After coupling with the ANG peptide and RVG peptide, the P_app value ratio of AP-BL increased compared to NMN, while the P_app value ratio of BL-AP decreased.

The P_app values of NMN-lips were similar to those of NMN, and there was no significant difference compared to the NMN group (AP-BL: p = 0.7565, BL-AP: p = 0.4952). Compared with the NMN group, ANG-NMN-lips and RVG-NMN-lips showed a highly significant difference (AP-BL: p = 0.0016, BL-AP: p < 0.0001) and a significant difference (AP-BL: p = 0.0424, BL-AP: p = 0.0369), respectively (Figure 7). Coupling with the ANG peptide increased the P_app value of NMN by one order of magnitude, indicating that the ANG peptide loading improved the NMN transport capacity, resulting in a higher NMN permeability coefficient and better transport ability.

3.7. Result of Intracellular Accumulation and Recovery

The transport solution samples in the transport and efflux experiments as well as the bidirectional transmembrane transport experiments and time- and concentration-dependency experiments were measured by the HPLC method. The result shows that there is no chromatographic peak of NMN in the chromatogram, indicating that NMN was not detected within the detection limit of the instrument, and the recovery rate of NMN in the transmembrane transport sample is also good. Therefore, it is determined that NMN is not present in the cells.

4. Discussion

Drug delivery to a specific site of action is the key to the effectiveness of the drug, and the biological barrier during drug delivery may prevent the accumulation of drugs at the site of disease, thus limiting the effective response to drug treatment. The blood–brain barrier is one of the biological barriers of the human body. The blood–brain barrier can prevent some substances (mostly harmful) from entering the brain tissue from the blood. Drugs need to cross the blood–brain barrier to exert their therapeutic effects and treat brain diseases.

NMN is widely present in natural animals and plants and has a positive effect on delaying aging and inhibiting the development of brain diseases. At present, there have been extensive studies on NMN in the academic community, but they have mainly focused on component analysis, quality control, and pharmacological activity research. There is still a blank in the research on crossing the BBB and the transport mechanism of NMN. Therefore, this study conducted a preliminary investigation into the transport mechanism of NMN and attempted to explore ways to increase NMN’s passage through the blood–brain barrier.

There are two methods for evaluating drug properties across the blood–brain barrier: in vivo and in vitro experiments. Compared with in vivo experiments, in vitro experiments have shorter experimental cycles and lower costs and can be used for high-throughput screening and to identify and study individual molecular mechanisms of blood–brain barrier transport [30]. We constructed an in vitro cell monolayer model evaluating across the BBB using hCMEC/D3 cells and proved the success of the cell monolayer model’s establishment through two methods.

A drug substance is considered to be poorly absorbed (0–20%) when P_app is lower than 1 × 10⁻⁶, moderately absorbed (20–70%) when P_app is between 1 and 10 × 10⁻⁶, and well absorbed (70–100%) when P_app is higher than 10 × 10⁻⁶ [31]. We investigated the capability and mechanism of NMN across the BBB using the cell monolayer model. The results of the transport and efflux experiments of NMN showed that the P_app value was between 1 and 10 × 10⁻⁶, indicating that the transmembrane transport ability of NMN is moderate. The main pathways for drugs to cross the BBB are passive transport, active transport, and endocytosis. Passive transport involves the transport from the bloodstream to the brain through hydrophilic pores on the cell membrane, known as paracellular pathways, under the influence of concentration differences. Active transport includes both transporter protein and receptor-mediated pathways, in which peptides and proteins bind to specific receptors, initiating the internalization of ligands and receptors into cells [16,32,33]. An important difference between these two pathways is that active transport routes mostly require the participation and support of carriers and motivation. Studies have shown that there is no significant difference in the P_app ratio between A-B/B-A and B-A/A-B, and it is less than 2, indicating that the two directions of transport are not concentration-dependent or saturated, indicating pure passive transport [34]. In this study, we found that the ratio of P_app values of NMN is about one, with no signs of carrier participation, thus we preliminarily inferred that the transport mechanism of NMN is passive transport.

On this basis, time- and concentration-dependency bidirectional transmembrane transport experiments with NMN were conducted. The time-dependency experimental results of AP-BL and BL-AP both showed a trend of first increasing and then stabilizing at the receiving end, indicating that saturation has been reached, and 120 min is the optimal administration time. The concentration-dependency experimental results of AP-BL and BL-AP both showed that the number of substances transported by NMN was concentration-dependent, and the higher the concentration, the higher the number of substances transported by NMN. The results of the transport percentage indicate that, when the concentration of NMN is greater than 0.5 mmol/L, the change in transport percentage is small, indicating that saturation has been reached when 0.5 mmol/L is administered. Therefore, the subsequent experimental administration concentration should be less than 0.5 mmol/L. This result further proves that the transmembrane transport mode of NMN is passive transport relying solely on concentration differences.

Given that the transport capacity of NMN was moderate, we modified the dosage form of NMN to prepare liposomes coupled with targeted peptides to enhance the transmembrane transport capability of NMN. After the successful preparation of liposomes, the particle size of KB-lips and NMN-lips is mainly concentrated around 100 nm, while the particle size of ANG-NMN-lips and RVG-NMN-lips is mainly concentrated around 130 nm. Nanoformulations with a particle size range of 100–200 nm are more suitable for intravenous administration, which can reduce the absorption of phagocytic cells and improve drug exposure time by prolonging blood circulation time [35]. The liposomes prepared in this study have suitable particle sizes and meet the requirements for intravenous injection, providing the possibility of developing NMN as an injection. Comparative experiments on the transport capacity of NMN liposomes and NMN showed that, compared with NMN, the P_app ratio of AP-BL increased after coupling with targeted peptides, while the P_app ratio of BL-AP decreased. The P_app values of NMN-lips and NMN are relatively similar, indicating that the transmittance of NMN cannot be increased. After coupling with the ANG peptide and RVG peptide, only ANG-NMN-lips increased the permeability coefficient of NMN to 10⁻⁵, and compared with the P_app value of NMN, ANG-NMN-lips extremely significantly improved the transport capacity of NMN, while RVG-NMN-lips significantly improved the transport capacity of NMN, indicating that ANG-NMN-lips makes it easier for NMN to pass through the BBB and increase its drug content in the brain, providing the theoretical basis for subsequent clinical studies.

Recent research has shown that developing a biocompatible polydopamine (PDA) platform and covalently attaching lactoferrin ligands to PDA spheres can improve the BBB transport efficiency of NMN through the exocytosis transmembrane pathway [36]. Compared with this study, the results of both experiments indicate that NMN can pass through the BBB and increase its ability to cross the BBB. The difference is that one is an in vivo experiment and the other is an in vitro experiment. This experiment not only has the advantages of simple and time-saving preparation methods but also has the advantage of being able to load larger amounts of drugs into the body and target the brain. Previous studies have confirmed that NMN is specifically transported into small intestinal cells by Slc12a8 transporters, which increase NAD+ levels in various organs and tissues of the body as they circulate through the bloodstream [37]. However, NMN is prone to degradation after reaching a certain level of moisture and temperature. At present, most NMNs on the market are capsules and tablets. After taking NMN capsules or tablets, most of them are degraded in the stomach, and only a small portion of NMN reaches the small intestine. The results of this study indicate that making NMN into an injection, transporting it directly to the brain through the blood, and improving the ability of NMN to pass through the BBB and delaying its release rate through NMN-loaded liposomes coupled with targeted peptides reduces NMN loss and provides the possibility of acting on the brain to a greater extent.

5. Conclusions

This is the first study on the ability and transport mechanism of NMN through an in vitro monolayer BBB model constructed based on the hCMEC/D3 cell line. Based on this model, it was found that NMN has moderate transmembrane transport ability and a passive transport mechanism. Moreover, NMN was successfully loaded into ANG-NMN-lips. Compared with the transmembrane transport ability of NMN, the transmembrane transport ability of ANG-NMN-lips is significantly improved and can be used to target the brain through intravenous injection. This study provided a theoretical basis for the subsequent clinical application of NMN and is of great importance for the development of new drugs for NMN in the central nervous system and related research on the targeted delivery of NMN.