1. pLAPs continuously activate insulin signaling pathway and promotes glucose uptake by sustained-release of LA in cells
The pLAPs were prepared by a nanoprecipitation-mediated polymerization and characterized by morphology, size, zeta potential, molecular weight and the proton nuclear magnetic resonance (Supplementary Figs. 1 and 2). Benefiting from the polymerization, the pLAPs exhibited robust dilution and serum stability, and dissociated only in PBS with glutathione (GSH, 2 ~ 10 mM) that mimicking the redox level in the insulin target cells (Supplementary Fig. 3). To investigate whether the self-polymerization prolonged the intracellular retention time of LA, three insulin target cells (3T3-L1, L6, and L02) were incubated with the conditioned medium containing 100 µg/mL pLAPs. After 1 h incubation, the medium containing pLAPs was removed and the pLAPs-free medium was added. At predetermined time points, the cells were lysed and the lysate supernatant was collected to detect the content of intracellular LA by high-performance liquid chromatography (HPLC). LA was set as a control. As plotted in Fig. 2a-c, in LA-treated cells, the LA content maintained a high level for the first 10 min, then rapidly decreased and was almost undetectable after 360 min. By comparison, in pLAPs treated cells, the LA content gave a time-dependent increase. After reaching the peak at 60 min, it remained at the high level over 360 min. We rationalized that the maintenance of the high LA level would be attributed to the balance of its dissociation and metabolism. In other words, the self-polymerization indeed significantly prolonged the intracellular retention time of LA. It should be pointed out that the self-polymerization mediated prolongation of LA intracellular time cannot be achieved by loading LA in any nanocarriers due to the unavoidance of the rapid diffusion release.
Before investigating the effect of prolonged intracellular retention of LA on the activation of insulin signaling pathway, the 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays were performed to evaluate the nontoxic concentration range of pLAPs on target cells (0 ~ 250 µg/mL, Supplementary Fig. 4). The activation of insulin signaling pathway was achieved by incubating cells with palmitic acid (PA) and LA or pLAPs for 1 h before the insulin stimulation (See Supporting Information for details). PA is a classic insulin resistance inducer, which can effectively inactivate the insulin signaling pathway by inhibiting the expression of p-IRS1 and p-Akt, two key insulin signaling molecules. As shown in Fig. 2d and Supplementary Fig. 5, the LA treatment increased only the expression of p-IRS1 and p-Akt in the first 10 min. By contrast, pLAPs caused no change of the expression of p-IRS1 and p-Akt within the first 30 min, while gave a significant increase at 60 min and maintained the high level within 60 ~ 360 min, consistent with the trend in intracellular LA content variation of the pLAPs group. As a result of the persistent activation of insulin signaling pathway, 2-(N-(7-Nitrobenz-2-oxa-1,3-diazol-4-yl)Amino)-2-Deoxyglucose (2-NBDG), a deoxyglucose analogue used for direct monitoring of glucose uptake by living cells, showed also a remarkable uptake level within 60 ~ 360 min (Fig. 2e-g).
To confirm the achievement of pLAPs’ above effects by dissociating into LA in the cells, the GSH depletion agent diethyl maleate (DEM) was employed to inhibit the intracellular dissociation of pLAPs (Fig. 2a-c, pink line). As shown in Fig. 2h-n, the block of the pLAPs dissociation led to the inactivation of insulin signaling pathway and the disappearance of 2-NBDG uptake, suggesting that pLAPs activates insulin signaling pathway and promotes glucose uptake by dissociating into LA rather than in the form of nanoparticles.
2. pLAPs hold good absorption, long in vivo half-life and efficient accumulation inT2DM target organs
The absorption effect of pLAPs was first investigated in vitro via the mucus penetration, epithelial cellular uptake and epithelial penetration, the major barriers of oral drug delivery. Rifampicin, a drug with oral bioavailability ~ 95%, was set as a positive control (See Supporting Information for details). As shown in Fig. 3a-f, the mucus penetration rate of pLAPs was ~ 61.0%, over 7/10 of rifampicin (~ 84.0%). Moreover, the cellular uptake efficiency and apparent permeability coefficient (Papp) of pLAPs reached ~ 71.8% and 2.6×10− 6 cm s− 1, respectively, data comparable to rifampicin (~ 77.4% and 2.8×10− 6 cm s− 1), indicating the good intestinal absorption of pLAPs. The good absorption was further evaluated by an ex vivo intestinal permeability model (Fig. 3g). Briefly, the pLAPs or rifampicin was injected into the lumen (donor chamber) before ligation, followed by an incubation in the Krebs-Ringer (K-R) buffer (acceptor chamber) at 37°C. As shown in Fig. 3h, as the drug content stabilized in the acceptor chamber, the small intestine absorption ratio of pLAPs was ~ 58.0%, up to 2/3 of rifampicin (~ 86.0%).
We hypothesize that the good absorption is correlated to the polydisulfide backbone of pLAPs, which enables dynamic covalent exchange with the thiols of mucin, epithelial cell membrane protein and lysosomal membrane so as to mediate the nanoparticles to penetrate the absorption barriers24,25. To check the hypothesis, 5,5’-dithiobis(2-nitrobenzoic acid) (DTNB), a popular thiol-reactive agent, was employed to block the thiol groups. As shown in Fig. 3i,j, just a 30 min blocking, the mucus penetration and cellular uptake efficiency of pLAPs decreased respectively by ~ 68.8% and ~ 31.1%. The lysosomal escape behavior of pLAPs was investigated by fluorescence colocalization of Coumarin 6-labeled pLAPs (Cou@pLAPs) and LysoTracker red in Caco-2 cells via confocal laser scanning microscope (CLSM). As shown in Supplementary Fig. 6, the Cou@pLAPs were colocalized with LysoTracker at 1 h, indicating the lysosomal entrapment. As the incubation time was prolonged from 1 to 8 h, the Cou@pLAPs and LysoTracker fluorescence gave less colocalization and the colocalization Pearson’s correlation coefficient (Rr) decreased dramatically from 0.72 to 0.31, indicating the escape of pLAPs from the lysosomes. In contrast, after the blocking of thiols on the lysosomal membrane, the Cou@pLAPs were colocalized with LysoTracker over 8 h, and the Rr showed no significant change.
The oral terminal half-life (T1/2) of pLAPs was studied on Sprague Dawley (SD) rats via oral gavage. At predetermined time points, the blood samples were collected and treated with NaOH, then the dissociated LA was extracted and measured to reflect the plasma concentration of pLAPs. As shown in Fig. 3k, in LA monomer-treated SD rats, the T1/2 was 0.5 ± 0.1 h, data comparable to the previous reports. To our delight, in the pLAPs-treated SD rats, the T1/2 was up to 23.2 ± 1.6 h, ~ 46 times higher than that of LA. We rationalized that the long T1/2 of pLAPs would be ascribed to its negative surface charge and ~ 30,000 molecular weight, which can decrease plasma protein binding and retard the glomerular filtration. As a result of the good absorption and prolonged T1/2, the area under the 72 h of plasma concentration curve (AUC0 ~ 72 h) of pLAPs was up to 3761.8 ± 55.9 h*µg/mL, ~ 23 times higher than that of LA, indicating that the self-polymerization does remarkably improve the oral pharmacokinetics of LA.
The biodistribution of pLAPs was evaluated on C57BL/6 mice through oral administration of 1,19-dioctadecyl-3,3,39,39-tetramethylindodica-rbocyanine perchlorate (Did)-labeled pLAPs (Did@pLAPs). At predetermined time points, mice were euthanized to harvest the major and insulin target organs, and the ex vivo fluorescence images were detected using an in vivo imaging system (IVIS). As shown in Fig. 3l,m, the ex vivo IVIS images revealed that the fluorescence signals were mainly observed in the liver, heart and kidney. Besides the distribution in the above main organs, one can find that the pancreas (insulin secretion organ) and fat and skeletal muscle (insulin effector organs) gave also strong fluorescence signals, indicating the T2DM target organs-efficient accumulation of pLAPs.
3. pLAPs exert strong efficacy in the murine model of T2DM
The oral hypoglycemic effect was assessed on the T2DM model of the 8-week-old male db/db mice by single and long-term administrations of pLAPs at the doses of 25, 50 and 100 mg/kg or LA at 100 mg/kg (Fig. 4a,b). The db/db mice treated with saline and metformin (120 mg/kg, equivalent therapeutic dose for T2DM patients) were set as the negative control and positive control, respectively, and db/m mice treated with saline was set as the normal control. As shown in Fig. 4c, LA caused no change of the blood glucose level as expected. Metformin gave a glucose decrease during 2 ~ 24 h, with a lowest level of 10.7 mmol/L at ~ 6 h. To our delight, the pLAPs showed hypoglycemic effects at all three doses, and the best one occurred in the 100 mg/kg group, where the hypoglycemic effect was maintained in a period of 2 ~ 72 h and the blood glucose level was as low as 8.4 mmol/L, 21.5% lower than that in the metformin group.
Due to the nearly 3-day hypoglycemic effect of a single administration, the pLAPs were administered every three days in the long-term evaluation, while the LA and metformin were administered daily. As shown in Fig. 4d, LA showed still no hypoglycemic effect. Metformin stabilized the blood glucose at ~ 12.7 mmol/L. By contrast, pLAPs administrated even every three days showed an excellent hypoglycemic effect, with the blood glucose stable at a low level of ~ 9.4 mmol/L (100 mg/kg), much lower than that in the metformin group and even close to the normal control.
To confirm the hypoglycemic effect of pLAPs, all mice were orally administrated with glucose (2 g/kg) on day 27 and intraperitoneally injected with recombinant human insulin (1 U/kg) on day 29, and the AUC0 ~ 120 min of oral glucose tolerance test (OGTT) and insulin tolerance test (ITT) were detected to evaluate glucose tolerance and insulin resistance, respectively. As shown in Fig. 4e, the db/db mice showed a significant AUC0 ~ 120 min increase of OGTT and ITT compared to the db/m mice, suggesting a seriously impaired glucose tolerance and insulin resistance. LA caused no improvement of the AUC0 ~ 120 min of OGTT and ITT, consistent with the hypoglycemic result. Metformin showed a decrease of 21.4% for OGTT and 27.1% for ITT, comparable to the other reports26. By comparison, pLAPs decreased the AUC0 ~ 120 min of OGTT by up to 39.5% and ITT by up to 48.3%, a comparable level to the normal control, implying the excellent improvement in glucose tolerance and insulin resistance. In addition, the improvement of pLAPs on insulin resistance was further evidenced by serum insulin and insulin resistance index (HOMA-IR) assays, in which the serum insulin level and HOMA-IR were decreased by 39.4% and 72.5%, respectively (Fig. 4f,g). Notably, one can find that the pLAPs reduced the weight of epididymal and body by 34.6% and 25.6%, respectively, while metformin caused no any weight loss (Fig. 4h,i), indicating the potential of pLAPs to reduce the risk factor of T2DM through weight loss. One can also find the weight loss of pLAPs was a little bit weaker than that of LA, possibly due to the fact that pLAPs were mainly distributed in insulin target organs and failed to block AMPK pathway in the hypothalamus to inhibit food intake27,28.
4. Besides the activation of classical IRS1/Akt signaling pathway, pLAPs exert hypoglycemic effect by promoting the normalization of pancreatic structure and secretion function
In order to confirm the activation of pLAPs on the classical IRS1/Akt signaling pathway in vivo, all mice were euthanized after 30-day treatment, and the expression of p-IRS1 and p-Akt in liver and skeletal muscle were detected by Western blotting. Consistent with the in vitro results, pLAPs showed a significant increase in the expression of p-IRS1 and p-Akt, and, as a result, the levels of glucose transporter 4 (GLUT4, a key protein that promotes glucose transmembrane mobility) and glycogen deposits were also remarkably increased (Supplementary Fig. 7).
In addition to the inhibition of classical IRS1/Akt signaling pathway, pancreatic damage caused by inflammation and oxidative stress is the other major factor in the progression of T2DM29,30. Considering the strong antioxidant and anti-inflammatory properties of pLAPs, we evaluated further its impact on pancreas lesions by hematoxylin and eosin (H&E), TUNEL and insulin immunohistochemical staining. As shown in Supplementary Fig. 8a-d, metformin had no improvement on pancreatic damage. To our delight, pLAPs, at the dose of 50 mg/kg, showed a significant improvement in the related indicators. When the dose was increased to 100 mg/kg, a much stronger improvement was achieved, where the size of islets was reduced by 70.9%, the number of apoptotic β-cells was reduced by 70.1%, and the expression of insulin was increased by 159.3%. Notably, one can find that all three levels were close to the normal control, indicating a very good improvement of pLAPs on pancreatic damage.
To confirm the improvement of pLAPs on the pancreas injury by antioxidation and anti-inflammatory, the oxidative stress marker malondialdehyde (MDA), antioxidant markers catalase (CAT), superoxide dismutase (SOD), and glutathione peroxide (GSH-Px), and pro-inflammatory cytokines tumor necrosis factor α (TNF-α), interleukin-1β (IL-1β) and interleukin-6 (IL-6) were assessed. As shown in Supplementary Fig. 8e-h, pLAPs gave a dose-dependent decrease on the level of oxidative stress marker and an increase on the activities of antioxidant markers, respectively, with the high-dose group (100 mg/kg) approaching to the normal control. Consistent with antioxidation results, pLAPs exhibited also a strong inhibitory effect on inflammation (Supplementary Fig. 8i-k). We note in passing that LA had no therapeutic effect on pancreatic injury, and its inhibition on oxidative stress and inflammation was also limited. This would be because LA metabolizes too fast and cannot be accumulated effectively in pancreatic tissue.
5. pLAPs show a much stronger therapeutic efficacy for neuropathy than LA and reduces blood glucose in the advanced stage of diabetes
The treatment of pLAPs on neuropathy was evaluated by using the 20-week-old male db/db mice as a DPN model. Consistent with the frequency of administration in the long-term hypoglycemic evaluation, pLAPs were orally administrated every three days for 30 consecutive days at a dose of 100 mg/kg, and once-daily oral administration of LA (100 mg/kg, equivalent therapeutic dose for clinical patients with DPN) was set as the positive control. Considering the oxidative stress pathogenesis of neuropathy, we first investigated the changes of oxidative stress and inflammation-related indicators at the end of treatment. As shown in Fig. 5a,b, pLAPs gave a significant decrease on the content of MDA and an increase on the activities of CAT, SOD and GSH-Px in serum and sciatic nerve (SN). To our delight, all these indicators showed no difference compared to the normal control (P > 0.05), implying the complete improvement of redox imbalance. Consistent with the improvement effect of oxidative stress indicators, the pro-inflammatory cytokines (TNF-α, IL-1β and IL-6) and inflammatory signaling molecules (p-IκBα and p-NF-κB) restored also to the normal control levels (P > 0.05, Fig. 5c-i). By contrast, LA, even with administration daily, had still an enormous gap in the levels of the above indicators compared to the normal control. These results indicated that by transforming LA to the stabilized nanoparticles (pLAPs), a superstrong antioxidant and anti-inflammatory capacity was achieved.
As a result of the performance in antioxidant and anti-inflammatory, the mice treated by pLAPs showed an excellent improvement in behavior and nerve conduction velocity, where the hot-plate and tail-flick latencies were shortened by 40.4% and 31.1%, respectively (Fig. 5j,k), and motor nerve conduction velocity (MCV) and sensory nerve conduction velocity (SCV) were increased by 35.2% and 36.7%, respectively (Fig. 5l,m). At the same time, the myelin sheath of SN was significantly regenerated (Fig. 5n). By contrast, the latencies of hot-plate and tail-flick in LA treated mice were reduced by only 13.5% and 14.2%, respectively, and the MCV and SCV were increased by only 12.9% and 19.7%, respectively.
As known, the 20-week-old db/db mice, equivalent to patients with T2DM more than 10 years, were generally used as a model for advanced diabetes. In the advanced stage of diabetes, the blood glucose lowering is difficult to achieve owing to the proliferation cessation and loss of secretory function of β-cells, as well as the aggravation of insulin resistance caused by long-term hyperglycemia. To our delight, pLAPs showed also a good and stable hypoglycemic effect in the rodent model of advanced diabetes, where a stable blood glucose occurred on 6th day, with a low level of ~ 16.3 mmol/L (Fig. 5o).
6. pLAPs exhibit excellent therapeutic effect on diabetes microvascular and macrovascular disease
As known, the lesion site of DPN locates in the peripheral nerves with a single structure (composed only of neurons and glial cells)31,32. LA easily spreads to nerves and is directly taken up by cells to exert systemic antioxidant effects. Therefore, LA is approved in clinic for the DPN treatment in despite of its fast metabolism. On the contrary, the diabetic angiopathy locates mainly in the complex organs, such as heart, kidney and retina. LA hardly accumulates in the effector cells of these organs. Although driven also by high glucose-mediated oxidative stress, the angiopathy cannot be treated with LA. Encouraged by its much stronger antioxidant and anti-inflammatory capacity over LA and effective enrichment in T2DM target tissues, we tried further the potential of pLAPs in the treatment of angiopathy. The diabetic nephropathy (DN) and diabetic cardiomyopathy (DC) were employed as models for microvascular disease and macrovascular disease, respectively. Consistent with the results in the DPN treatment, pLAPs restored the oxidative stress and inflammation-related indicators to normal control levels in both models (Supplementary Fig. 9) and showed also a good and stable hypoglycemic effect (Supplementary Fig. 10).
The therapeutic effect of DN was evaluated by measuring the renal function indexes [urine albumin-to-creatinine ratio (UACR), serum creatinine (Cre), and blood urea nitrogen (BUN)] and glomerular pathological changes [glomerular sizes, glomerular basement membrane (GBM) thickness, mesangial matrix expression, and width and number of foot processes]. As shown in Fig. 6, compared to mice treated with saline, the pLAPs treated mice showed a decrease on all three renal function indexes, with the levels of UACR, Scr and BUN reduced by 71.3%, 52.6% and 50.9%, respectively (Fig. 6a-c). The glomerular pathological results showed that pLAPs reduced the glomerular size by 64.4%, GBM thickness by 18.3%, mesangial matrix expression by 48.4%, the width of foot processes by 28.1%, and increased the number of foot processes by 64.2%, respectively (Fig. 6d-i). The therapeutic effect of DC was evaluated by measuring the cardiac function indexes [left ventricular ejection fraction (LVEF), left ventricular fractional shortening (LVFS), left ventricular end-systolic diameter (LVIDs), and left ventricular end-diastolic diameter (LVIDd)], pathological changes (ventricular myocytes size and fibrotic area), and myocardial injury markers [natriuretic peptide type A (ANP), collagen type I alpha 1 (Col1α1), matrix metallopeptidase 2 (MMP2) and transforming growth factor beta 1 (TGF-β1)]. As shown in Fig. 6j-n, compared to mice treated with saline, the pLAPs treated mice showed an increase of 82.9% for LVEF and 46.9% for LVFS, and a decrease of 53.6% for LVIDs and 32.4% for LVIDd. At the same time, the ventricular myocytes size, fibrotic area and myocardial injury markers levels were significantly decreased (Fig. 6o-t). In particular, one can find that the above indicators of DC and DN showed no significant difference compared to the normal control, suggesting that the mice returned to normal after the pLAPs treatment. As a contrary, the LA treatment only provided a very limited therapeutic effect from the slight improvement of the indicators. Taken together, pLAPs hold a good clinical potential in the treatment of both angiopathy and neuropathy.
7. pLAPs improve spontaneous T2DM with DPN in rhesus monkey
Encouraged by the outstanding outcomes in rodents, we challenged further the treatment of pLAPs for the spontaneous T2DM rhesus monkey with the complication of DPN since it very closely mimicked human T2DM pathological properties. Concretely, 5 spontaneous T2DM rhesus monkeys, which conform to either the slowing down of any SCV or prolonged latency by detecting SCV of the sural nerve, median nerve, and other nerves over 6 months using neuroelectromyography, were randomly divided into 2 groups. Three monkeys were given pLAPs (10 mg/kg/d) by nasogastric gavage for 8 weeks, and the other two were given placebo as the negative control. It should be pointed out that, for meeting the requirement of T2DM combined with DPN, the selected five animals are elderly diabetes monkeys aged 18–21 years (analogous to approximately 70-year-old humans)33,34. At this stage, even the classic insulin-sensitizing drugs such as metformin, sulfonylureas, or GLP-1RA would lose their hypoglycemic effect caused by islet function failure35,36,37,38,39. Such situation occurred also in our case. Although the fasting plasma glucose (FPG) of the pLAPs treated monkeys gave a downtrend compared to monkeys treated with placebo, there was no significant difference compared to baseline (Table 1). Considering that the FPG detection in this stage is no longer suitable for evaluating the true effectiveness of drugs on T2DM, the HOMA-IR, a unique indicator reflecting the insulin sensitivity without affection by the disease-developing stage, was employed to evaluate the therapeutic effect of T2DM. The lower the HOMA-IR is, the stronger the hypoglycemic effect of drugs is. As shown in Table 1, the placebo treated monkeys showed no change of the HOMA-IR during the treatment. To our delight, the HOMA-IR of monkeys treated by pLAPs gave a continuous decrease from 7.41 to 2.02, a value located in the range of normal human (< 2.69)40, verifying undoubtedly the excellent therapeutic efficacy of pLAPs for T2DM.
Table 1
General glucose metabolism indicators of the T2DM rhesus monkey with DPN following once daily oral administration of pLAPs (10 mg/kg) or placebo for 8 weeks.
Indicators | Group | Baseline | Day 14 | Day 28 | Day 42 | Day 56 |
FPG (mmol/L) | Placebo | 4.93 ± 0.08 | 5.18 ± 0.60 | 4.91 ± 0.08 | 4.79 ± 0.25 | 4.80 ± 0.16 a |
pLAPs | 6.97 ± 2.58 | 6.88 ± 2.81 | 6.84 ± 1.95 | 6.83 ± 2.39 | 6.43 ± 2.48 a |
FPI (µU/mL) | Placebo | 14.81 ± 6.16 | 22.76 ± 16.59 | 12.74 ± 2.36 | 18.53 ± 5.83 | 17.30 ± 10.22 a |
pLAPs | 22.77 ± 7.83 | 13.03 ± 10.22 | 11.56 ± 7.17 | 9.46 ± 11.55 | 7.86 ± 7.61 b |
HOMA-IR | Placebo | 3.23 ± 1.30 | 5.01 ± 3.21 | 2.78 ± 0.47 | 3.91 ± 1.04 | 3.65 ± 2.05 a |
pLAPs | 7.41 ± 4.51 | 3.75 ± 2.43 | 3.25 ± 1.67 | 2.51 ± 2.72 | 2.02 ± 1.58 b |
Note: HOMA-IR = FPG × FPI/22.5. |
Statistical differences were analyzed by two-tailed Student’s t-test. aP > 0.05, bP < 0.05, compared with baseline. |
FPG: fasting plasma glucose; FPI: fasting plasma insulin; HOMA-IR: insulin resistance index. |
The therapeutic effect for DPN was assessed by measuring the nerve conduction velocity (NCV) every 4 weeks, and the data were shown in Table 2. During the treatment, the placebo caused no change of the NCV of abnormal nerves compared to that at baseline (P > 0.05), while the NCV of pLAPs treated group showed a continuous increase from 37.6 ± 3.4 m/s to 41.2 ± 3.2 m/s. By directly comparing the velocity change on day 28 and day 56, one can find that the placebo’s velocity changes were − 0.3 ± 0.8 m/s and − 0.2 ± 1.1 m/s, respectively, while the corresponding values of pLAPs gave a significant improvement of 2.4 ± 2.8 m/s and 3.6 ± 2.5 m/s, suggesting the significant improvement of pLAPs on DPN. To our knowledge, this is the first hypoglycemic drug with the function of alleviating neuropathy.
Table 2. NCV and velocity change of the T2DM rhesus monkey with DPN following once daily oral administration of pLAPs (10 mg/kg) or placebo for 8 weeks.
Group
|
NCV (m/s)
|
|
Velocity change (m/s)
|
Baseline
|
Day 28
|
Day 56
|
|
Day 28
|
Day 56
|
Placebo
|
38.1 ± 4.9
|
37.8 ± 4.8 a
|
37.9 ± 4.8 a
|
|
-0.3 ± 0.8
|
-0.2 ± 1.1
|
pLAPs
|
37.6 ± 3.4
|
40.0 ± 3.6 a
|
41.2 ± 3.2 b
|
|
2.4 ± 2.8 c
|
3.6 ± 2.5 c
|
Note: Statistical differences were analyzed by two-tailed Student’s t-test. aP > 0.05, bP < 0.05, compared with baseline. cP < 0.05, compared with placebo group.
NCV: nerve conduction velocity.
8. pLAPs hold good biosafety
As a chronic metabolic disease, T2DM requires long-term treatment with hypoglycemic drugs. Therefore, the drug safety is particularly important. Here, we evaluated the biosafety of pLAPs by acute and subacute toxicity tests. The acute toxicity was performed by single oral administration of 2000 mg/kg pLAPs into C57BL/6 mice, saline was used as a control. The results showed that the mice maintained healthy as confirmed by the unchanged behavior and nonweight loss over 2 weeks (Supplementary Fig. 11a). At the end of the observation, mice were euthanized and the main organs and blood were collected for the organs index, hematology profile and serum biochemistry analyses. It was found that no obvious changes in the organs index (Supplementary Fig. 11b), hematology profile (Supplementary Fig. 11c) and serum biochemical parameters (Supplementary Fig. 11d) were observed in mice treated with pLAPs.
The subacute toxicity was performed by oral administration of pLAPs (100 mg/kg/d) to C57BL/6 mice for 30 days. Saline was used as a control. The levels of hematology profile, plasma biochemical parameters and the pathological changes of main organs were checked after treatment. The results showed that the mice treated with pLAPs did not show significant differences in the hematology (Supplementary Fig. 12a) and plasma biochemical parameters (Supplementary Fig. 12b) compared with the control group. In addition, the H&E staining of the main organs, including the heart, liver, spleen, lung and kidney showed no abnormalities after pLAPs oral administration (Supplementary Fig. 13). Notably, thanks to the hypoglycemic mechanism of the insulin signaling pathway activation and pancreatic injury improvement, the mice treated with pLAPs showed no hypoglycemic reactions under the doses of both acute and subacute toxicity (Supplementary Fig. 14). By the way, during the 8 weeks of administering pLAPs (10 mg/kg/d) via nasogastric gavage to treat spontaneous T2DM rhesus monkey with DPN, we did not observe any abnormalities, including body weight, food intake (Supplementary Table 1), plasma biochemical parameters (Supplementary Table 2) and hematology profile after treatment (Supplementary Table 3). Overall, the above results suggested the good biosafety of pLAPs.