Potentiële osteo-inductieve effecten van hydroxyapatiet-nanodeeltjes op mesenchymale stamcellen door interactie met endotheelcellen

Resultaten

Karakterisering van HAP's

Zoals weergegeven in figuur 2, werden HAP's bereid met een bepaalde grootte en vorm. De diameters van de np20 met bijna bolvorm waren gemiddeld 20 nm, en de np80 was staafvormig met een gemiddelde lengte van 80 nm en een breedte van 20 nm. De m-HAP's waren ook bijna bolvormig en ongeveer 12 m in diameter. Alle deeltjes hadden een negatieve oppervlaktelading in L-DMEM. Er is gesuggereerd dat negatieve waarden van zeta-potentiaal een significant gunstig effect hebben op de aanhechting en proliferatie van botcellen, evenals op de directe botbinding en nieuwe botvorming [22, 23]. Deeltjes, die werden waargenomen in L-DMEM, hebben de neiging om te aggregeren in waterige systemen. Hun hydrodynamische grootte werd ook getest, wat ook een belangrijke factor kan zijn die hun biologische gedrag beïnvloedt.

Karakterisering van HAP's. TEM-microfoto's van a np20 en b np80 en SEM-microfoto van c m-HAP. d Karakterisering van HAP's (n = 6). Afkortingen:TEM transmissie-elektronenmicroscopie, SEM scanning elektronenmicroscopie, HA hydroxyapatiet, m-HAP micro-sized HAP-deeltjes

Indirecte toxiciteit van HAP's voor hWJ-MSC's

Om de indirecte toxiciteit van HAP's op hWJ-MSC's te beoordelen, werd de levensvatbaarheid van de cellen gemeten via MTS-assays. CM met 50 µg/ml HANP's zou de levensvatbaarheid van hWJ-MSC aanzienlijk kunnen stimuleren na 24 uur en 72 uur en vooral na 24 uur. CM met 100 µg/ml HANP's verminderde echter de levensvatbaarheid van de cellen met 15-20% in vergelijking met de controle na 72 uur. Bovendien stimuleerde CM met 25 µg/ml np20, maar niet np80, de levensvatbaarheid van de cellen na 24 uur. Deze verschijnselen bevestigden dat 50 µg/ml HANP's een subcytotoxische concentratie waren die in alle volgende experimenten werd gebruikt (Fig. 3).

Indirecte toxiciteit van HAP's voor hWJ-MSC's. De levensvatbaarheid van hWJ-MSC's die indirect zijn gekweekt met HAP's werd gemeten voor a 24 en b 72 uur. *P < 0,05; **P < 0,01 versus de controle. De controlegroep bestond uit cellen die waren geïncubeerd in CM zonder behandeling met HAP's, en de levensvatbaarheid van de cellen was genormaliseerd als een percentage van de controle. Afkortingen:HAP's hydroxyapatietdeeltjes, m-HAP micro-sized HAP-deeltjes, hWJ-MSC's menselijke navelstreng Uit gelei afgeleide mesenchymale stamcellen van Wharton

Indirect osteo-inductief effect van HAP's op hWJ-MSC's

Om de indirecte osteo-inductieve effecten van HANP's op hWJ-MSC's te identificeren, werd de expressie van osteogene genen beoordeeld door kwantitatieve RT-PCR-analyse. Het transcriptniveau voor runt-gerelateerde transcriptiefactor 2 (RUNX-2) in de HANP-groepen, met name np20, vertoonde een opmerkelijke toename van dag 7 tot dag 21 (figuur 4a). De genexpressie van type I collageen (Col I) in de np20-groep vertoonde een verbetering van dag 7 tot dag 21, terwijl de np80-groep een aanhoudende toename liet zien van dag 7 tot dag 14 (Fig. 4b). Osteocalcine (OCN) mRNA was duidelijk opgereguleerd in de HANP-groepen op dag 14, wat wijst op een versnelde osteogenese (figuur 4c). De mRNA-niveaus van alkalische fosfatase (ALP) namen duidelijk toe in de HANP-groepen (figuur 4d), wat wijst op de osteogene differentiatie van hWJ-MSC's. De m-HAP-groep vertoonde echter beperkte veranderingen met betrekking tot de niveaus van deze drie osteogene genen in vergelijking met de controlegroep (Fig. 4). Bovendien nam de expressie van pluripotentiemarkers, NANOG, OCT3/4 en SOX2, af in de HANPs-groepen in vergelijking met de controle (Fig. 4e-g), wat impliceert dat de hWJ-MSC's in de HANPs-groepen waren gedifferentieerd, vooral in de np20-groep. Vergelijkbare resultaten werden verkregen door Western-blot-analyse (Fig. 5c, d), wat aangeeft dat de np 20-groep indirect de expressie van RUNX-2 en COL I in hWJ-MSC's zou kunnen verbeteren.

Indirecte effecten van HAP's op de expressie van osteogene differentiatie-gerelateerde genen. een RUNX-2, b Kolom I, c OCN, d ALP, e NANOG, v OCT3/4, en g SOX2-genniveaus in hWJ-MSC's gekweekt met CM gedurende de aangegeven tijd. *P < 0,05; **P < 0,01; ***P < 0,001 versus controlegroep; ^& P < 0,05; ^&& P < 0,01; ^&&& P < 0,001 versus de m-HAP-groep; ^# P < 0,05; ^## P < 0,01 versus de np20-groep. Cellen die waren geïncubeerd in osteogeen medium zonder HAPs-behandeling werden gebruikt als de controlegroep. Afkortingen:HAP's hydroxyapatietdeeltjes, m-HAP micro-sized HAP-deeltjes, hWJ-MSC's menselijke navelstreng Wharton's gelei-afgeleide mesenchymale stamcellen, RUNX-2 runt-gerelateerde transcriptiefactor 2, Col I type I collageen, OCN osteocalcine, ALP alkalische fosfatase, SOX 2 SRY-gerelateerde HMG-box 2

Indirect effect van HAP's op extracellulaire calciumafzetting en ALP-activiteit. hWJ-MSC's werden 14 dagen in osteogeen medium geïncubeerd. een Extracellulaire calciumafzetting werd vervolgens zichtbaar gemaakt via ARS-kleuring; b ALP-activiteit van de hWJ-MSC's werd beoordeeld via ALP-kleuring, schaalbalken:200 m. c Western-blot-analyse gaf de expressie aan van RUNX-2 en COL I van hWJ-MSC's in osteogeen medium op dag 7. d Densitometrische metingen van RUNX-2 en COL I uit deel (c ). Cellen die waren geïncubeerd in osteogeen medium zonder HAPs-behandeling werden gebruikt als de controlegroep. *P < 0,05; **P < 0,01; ***P < 0,001 versus controlegroep; ^& P < 0,05; ^&&& P < 0,001 versus de m-HAP-groep; ^### P < 0,001 versus de np20-groep. Afkortingen:HAP's hydroxyapatietdeeltjes, m-HAP micro-sized HAP-deeltjes, hWJ-MSC's menselijke navelstreng Wharton's gelei-afgeleide mesenchymale stamcellen, ALP alkalische fosfatase

Om het indirecte osteo-inductieve effect van HANP's op hWJ-MSC's visueel te observeren, werden de cellen 14 dagen geïncubeerd met het aangegeven osteogene medium, gevolgd door ARS- en ALP-kleuring. Zoals getoond in Fig. 5a, b, werd een verhoogd aantal gemineraliseerde knobbeltjes en hogere ALP-activiteit van de hWJ-MSC's waargenomen in de HANP-groepen in vergelijking met de m-HAP- en controlegroepen. Bovendien vertoonde m-HAP, vergelijkbaar met de controle, beperkte effecten op de osteogene differentiatie van de hWJ-MSC's.

HAP's geactiveerde ERK1/2-signalering in hWJ-MSC's indirect samen gekweekt met HUVEC's

Om de effecten op de paracriene functie van HUVEC's door HAP's te onderzoeken, werden immunofluorescentie- en ELISA-assays gebruikt om het mogelijke eiwit te identificeren dat de osteogene differentiatie van de hWJ-MSC's bevordert. Zoals getoond in Fig. 6a-c, werd de intracellulaire en extracellulaire productie van HIF-1α aanzienlijk vergemakkelijkt door HANP's (vooral np20), terwijl er een beperkt effect was van m-HAP op de productie ervan.

HAP's activeerden ERK1/2-signalering in hWJ-MSC's die indirect samen met HUVEC's werden gekweekt. een Immunofluorescentie van HIF-1α uitgevoerd in HUVEC's behandeld met/zonder HAP's gedurende 18 uur. b De fluorescentie-intensiteit van HIF-1α uit deel (a ). c De extracellulaire concentratie van HIF-1α in het kweekmedium van HUVEC's behandeld met/zonder HAP's gedurende 18 uur werd gemeten via ELISA. Schaalbalken:20 μm. *P < 0,05; **P < 0,01; ***P < 0,001 versus controle; ^## P < 0,01; ^### P < 0,001 versus de np20-groep. Cellen zonder HAPs-behandeling werden gebruikt als de controlegroep. hWJ-MSC's werden 24 uur met CM behandeld. d Western-blot-analyse die activering van sleutelkinasen in de ERK1/2-routes aangeeft. e Densitometrische metingen van p-ERK1/2 van onderdeel (b ). **P < 0,01; ***P < 0,001 versus controle; ^## P < 0.01, ^### P <-0,001 versus 2-MeOE2( −) groep. Cellen geïncubeerd in CM zonder HAP-behandeling werden gebruikt als de controlegroep. Afkortingen:HAP's hydroxyapatietdeeltjes, m-HAP micro-sized HAP-deeltjes, hWJ-MSC's menselijke navelstreng Wharton's gelei-afgeleide mesenchymale stamcellen, HUVEC's menselijke endotheelcellen van de navelstrengader, ERK extracellulair signaal-gereguleerd kinase, HIF-1α hypoxie-induceerbare factor 1α

Om de differentiatiesignaleringsroute van de hWJ-MSC's geactiveerd door HIF-1α nauwkeuriger te begrijpen, hebben we de belangrijkste regulatoren van de ERK1/2-route onderzocht via Western-blot-analyse. Zoals getoond in Fig. 6d, e, terwijl de eiwitniveaus van totaal ERK1/2 ongewijzigd bleven, waren de p-ERK1/2-niveaus duidelijk verhoogd in de hWJ-MSC's gekweekt met HANP's, en dit was vooral het geval in de np20-groep. M-HAP had echter weinig effect op p-ERK1/2-niveaus in de hWJ-MSC's, vergelijkbaar met het effect op de HIF-1α-productie in HUVEC's. Belangrijk is dat de verhoogde niveaus van p-ERK1/2 in de hWJ-MSC's geactiveerd door HIF-1α kunnen worden geblokkeerd door 2-MeOE2, een specifieke HIF-1α-remmer, wat aangeeft dat HIF-1α stroomopwaarts van de ERK1/2-signaleringsroute functioneerde in hWJ-MSC's.

HIF-1α bevorderde osteogene differentiatie van hWJ-MSC's via de ERK1/2-route

Om te bepalen of HIF-1α noodzakelijk was voor de waargenomen stimulatie van hWJ-MSC osteogene differentiatie, werd een specifieke HIF-1α-remmer (2-MeOE2) op deze celculturen toegepast. Zoals getoond in Fig. 7, waren de gemineraliseerde matrixafzetting en ALP-activiteit van de met 2-MeOE2 (+) behandelde groep hWJ-MSC's gekweekt in osteogeen medium verzwakt, wat aangeeft dat HIF-1α onmisbaar was voor de osteogene differentiatie van de hWJ- MSC's. Op basis van deze resultaten hebben we de rol van de ERK1/2-route in de osteogene differentiatie van de hWJ-MSC's geactiveerd door HIF-1α verder onderzocht. De afzetting van gemineraliseerde matrix en ALP-activiteit in de hWJ-MSC's gekweekt met osteogeen medium werden onderdrukt na de toediening van PD98059, een specifieke MEK-remmer.

HIF-1α bevorderde osteogene differentiatie van hWJ-MSC's via de ERK1/2-route. hWJ-MSC's werden gedurende 14 dagen geïncubeerd in osteogeen medium met of zonder PD98059. een Extracellulaire calciumafzetting werd zichtbaar gemaakt via ARS-kleuring. b ALP-activiteit van hWJ-MSC's werd beoordeeld via ALP-kleuring. Schaalbalken:200 μm. c Quantitative analysis of extracellular calcium matrix. *P < 0,05; **P < 0,01; ***P < 0.001 versus control; ^# P < 0,05; ^### P  < 0.001 versus the np20 group. Cells incubated in osteogenic medium without HAPs and PD98059 treatment were used as the control group. Abbreviations:HAPs hydroxyapatite particles, m-HAP micro-sized HAP particles, hWJ-MSCs human umbilical cord Wharton’s jelly-derived mesenchymal stem cells, HUVECs human umbilical vein endothelial cells, ERK extracellular signal-regulated kinase, HIF-1α hypoxia-inducible factor 1α, ALP alkaline phosphatase

Two-Stage Cell-Lineage Models

To quantitatively reveal the intrinsic connection between the concentration of HIF and osteogenic differentiation of hWJ-MSCs, 0–4000 pg/mL of HIF-1α was used to treat eight groups of hWJ-MSCs. A two-stage cell-lineage mathematical model was used to analyze the proliferation, apoptosis, and osteogenic differentiation rates of these hWJ-MSCs treated with different concentrations of HIF-1α. As shown in Fig. 8a, fitting data were employed to obtain the simulated formula (\(\frac{d}{{d_{0} }} =\frac{1}{{0.14H^{2} - 0.43H + 1}}\)) and curve (blue curve), showing that the differentiation rate first increased and then decreased with the increase in HIF concentration. More specifically, the differentiation rate reached a peak at 1500 pg/mL HIF-1α.

Two-stage cell-lineage models. hWJ-MSCs were incubated in a defined concentration of HIF-1α for the indicated times. een Relative differentiation rates of hWJ-MSCs at different concentrations of HIF-1α, b relative ALP activity (differentiation degrees of hWJ-MSCs) and relative osteoblast cells on different culture days. c Three-dimensional surface of differentiation degree evolving with time and HIF-1α. Cell number with an initial seeding density of 1500 cells per well (in a 96-well plate), as well as d 0, e 375, and f 1500 pg/mL HIF-1α. Abbreviations:hWJ-MSCs human umbilical cord Wharton’s jelly-derived mesenchymal stem cells, HIF-1α hypoxia-inducible factor 1α, ALP alkaline phosphatase

According to our study, the concentrations of HIF-1α produced by HUVECs in the control, m-HAP, np80, and np20 groups were 240, 300, 325, and 375 pg/mL, respectively. The black square represents the differentiation rate promoted by 240, 300, 325, and 375 pg/mL HIF-1α, and this matches well with the simulated curve. In addition, the differentiation degrees of the hWJ-MSCs (relative ALP activity) treated with different concentrations of HIF-1α increased similarly increased with time. Therefore, in order to simplify the model, we consider their growth patterns to be similar or even identical. We found that the increase in ALP activity from Day 0 to Day 7 was proportional to the number of osteoblasts, and osteoblast cells reached their peak at the platform period. Therefore, after fitting the ALP activity with the relative osteoblasts cell number curve (osteoblasts cell number / maximum osteoblasts cell number), the maximum of ALP activity was predicted and relative ALP activity (ALP activity/ maximum of ALP activity), representing the differentiation degree, was acquired (Fig. 8b). Combining these two studies, the three-dimensional surface of the differentiation degree evolving with time and HIF-1α was obtained (Fig. 8c).

In order to estimate the optimal culture time, it was necessary to simulate cell numbers. After elucidating the differentiation rate under different concentrations of HIF-1α, we simulated the size of each population. The simulation utilized an initial seeding density of 1500 cells per well (in a 96-well plate) using different concentrations of HIF-1α. As shown in Fig. 8d–f, the experimental data (black square) match well with the simulated total cell numbers (blue curve), which is a sum of the number of hWJ-MSCs and osteoblasts, and this supports the ability of this model to predict the number of hWJ-MSCs and osteoblasts at any time point. Moreover, the osteoblast cell number approached the platform period at 21, 18, and 15 days with 0, 375, and 1500 pg/mL HIF-1α, respectively. This model provides the optimum culture time for guiding tissue engineering, and it also provides direct evidence that HIF-1α accelerates the osteogenic differentiation of hWJ-MSCs.

Discussie

With recent advances in nanobiomaterials, nano-based artificial bone substitutes have been an area of intense investigation. The accumulating evidence suggests that there are complex interactions between cells and nanobiomaterials due to their capacity to penetrate cell membranes and increase internal retention times [24, 25]. A previous study revealed that collagen/alginate nanofilms can adsorb onto the MSC membrane to activate intracellular signaling cascades and promote osteogenic differentiation [26]. Elegant experiments by Wu and his colleagues clearly demonstrated that TiO₂ nanotubes can improve vascularization and osteogenic differentiation by facilitating paracrine effects and cell junctions via EC-MSC interactions [27]. For the purpose of developing excellent candidates for bone tissue engineering, it is necessary to clarify the direct crosstalk between nano-based bone substitutes and cells implicated in bone repair as well as their indirect interactions. However, our current understanding of this is still limited. In the present study, we utilized an indirect co-culture model to further elucidate the biological effects of HANPs on MSCs in regard to the indirect interactions mediated by ECs.

Cytotoxicity is a primary issue for assessing the biocompatibility of any nanobiomaterial. Although our previous study found that HANPs did not directly influence the viability or apoptosis of hWJ-MSCs, they may still exert different impacts via the mediation of other cells [28]. Thus, it was necessary to evaluate the cytotoxic effects of HANPs on hWJ-MSCs mediated by HUVECs. Interestingly, after incubation in CM for 24 h and 72 h, hWJ-MSC viability was maintained and even elevated in the 0–50 µg/mL HANP groups, especially in the np20 group, indicating the existence of effector molecules in the CM. When the concentration of HANPs reached 100 µg/mL, they became cytotoxic to the hWJ-MSCs. However, 0–100 µg/mL m-HAP had no influence on hWJ-MSC viability (Fig. 3). Jiang et al. have shown that engineered nanoparticles of a particular size can have distinct endocytic routes and kinetics associated with altered downstream signaling involved in regulating target cell functions [29]. In our previous study, we showed that np20 and np80 were endocytosed by HUVECs, and this was followed by morphologic changes and the appearance of large vacuoles, indicating the activated state of the HUVECs. Additionally, np20, with their faster uptake speed and increased accumulation, might result in a stronger activation of HUVECs, possibly resulting in increased hWJ-MSC viability via paracrine signaling. Conversely, few m-HAPs can be endocytosed by HUVECs, and this might account for their limited influence on the metabolism of hWJ-MSCs [9].

To further explore the potential osteoinductive effect of activated HUVECs, a subcytotoxic dose of 50 µg/mL HAPs was used in subsequent studies. The CM collected from the activated HUVECs promoted extracellular calcium deposition, ALP activity, and osteogenic proteins expression in hWJ-MSCs, as well as the mRNA expression of osteogenic genes (Figs. 4, 5). Runx2, an essential transcription factor involved in specifying the osteoblast lineage [30], showed a substantial enhancement in the np20 group, indicating a strong osteoinductive effect on hWJ-MSCs (Fig. 4a and Fig. 5c, d). The np20 group demonstrated a 1.5-fold improvement in COLI expression at Day 7 (Figs. 4b, 5c, d) and a double increase at Day 14, which implied the presence of additional differentiated osteoblasts in the HANP-treated groups (Fig. 4b) [30]. OCN is a mature stage bone marker [31], and this gene showed a significant increase in the HANP groups at Day 14 (Fig. 4c), indicating that np20 and np80 can accelerate bone maturation compared to m-HAP. ALP is an early marker of osteoblast differentiation, and it obviously increased with culture time in each group, especially the np20 group, revealing that additional transformation occurred from MSCs to osteoblasts (Fig. 4d). Pluripotency markers, NANOG, OCT3/4, and SOX2 imply the capacity for differentiation [32]. As shown in Fig. 4e–g, the decreased expression in the genes of the HANP groups implied that most of the hWJ-MSCs in HAP groups had transformed into osteoblasts.

Our data demonstrated that the endocytosis of HANPs by HUVECs was associated with an improved osteogenic differentiation of hWJ-MSCs. However, the cause of this outcome is currently unclear. In terms of the paracrine function of HUVECs, we focused on soluble differentiation-inducing proteins in the supernatant of activated HUVECs. HIF-1α signaling is essential in coupling ossification and angiogenesis during bone regeneration [33, 34]. Heikal et al. reported that injured ECs secrete more HIF-1α even under normoxia conditions [35]. It has also been shown that exposure to HANPs inhibits the angiogenic ability of HUVECs [9]. Thus, we measured the concentrations of HIF-1α in the CM, and the results showed that the HIF-1α content increased in the HANP treatment groups compared to the m-HAP and control groups (Fig. 6a). To identify the role of HIF-1α in the osteogenic differentiation of hWJ-MSCs, we used 2-MeOE2, which is a specific HIF-1α inhibitor, was used. The decreased concentration of HIF-1α paralleled the impaired mineralized matrix deposition and ALP activity in these hWJ-MSCs, indicating that HANPs can promote the HIF-1α production of HUVECs to facilitate the osteogenesis of hWJ-MSCs (Fig. 7).

To properly apply HANPs for use in bone tissue engineering, it is necessary to gain further insights into the mechanisms by which HANPs promote the osteogenic differentiation of hWJ-MSCs mediated by HUVECs. The ERK1/2 pathway is downstream of HIF-1α [36] and is fundamental to the differentiation of MSCs [37]. In this work, the concentrations of HIF-1α in the CM coincide well with the p-ERK1/2 levels in the hWJ-MSCs (Fig. 6b, c). When 2-MeOE2 was applied, the p-ERK1/2 expression in the hWJ-MSCs failed to be activated, indicating that HIF-1α functioned upstream of ERK1/2 signaling. To directly address the role of ERK1/2 signaling in the osteogenic differentiation of hWJ-MSCs, PD98059, a specific MEK inhibitor, was used. The suppression of ERK1/2 signaling resulted in the lowest osteogenic differentiation of hWJ-MSCs. One possible reason for this occurrence is that the ERK1/2 pathway plays a key role in both HIF-1α signaling and in the apoptosis and proliferation signaling pathways, which could be responsible for the observed changes in osteogenic differentiation in these cells [38, 39]. Additionally, this could also be related to the presence of vascular endothelial growth factor (VEGF). VEGF is one of the downstream effectors of HIF-1α signaling [33], and it can also promote the osteogenic differentiation of MSCs via activation of the ERK1/2 pathway [37]. Our previous study found that np20 induced the production of VEGF in HUVECs [9]; therefore, it is possible that the suppression of the ERK1/2 pathway may result in inhibition of VEGF, which would lead to the decreased osteogenic differentiation of hWJ-MSCs. According to the available experimental results, we can summarize as follows. HANPs are able to more optimally process better direct [5] and indirect osteoinductive effects than m-HAPs. Compared to autogenous bone grafts and bone allografts, there is an extensive source of HANP and without secondary damage and potential immunogenicity. However, compared to m-HAPs, HANPs can suppress the angiogenic ability of HUVECs [9] and exhibit slight cytotoxicity in both a time- and dose-dependent fashion.

Recently, growing evidence has demonstrated the importance of HIF-1α in the bone regeneration. However, few studies have been able to quantitatively predict the MSC differentiation rate under specific initial conditions, such as the HIF-1α concentration. Taking cell proliferation, apoptosis, and osteogenic differentiation into account, we present a mathematical model that combines a two-stage cell lineage with HIF-1α that is highly correlated with our experimental data. By fitting the differentiation rate of hWJ-MSCs in 0–4000 pg/mL HIF-1α, we acquired the equations for describing the differentiation rate, HIF-1α concentration, and time. As shown in Fig. 8d, this model can depict the cell number map under different HIF-1α concentrations, so that it is possible to explore the intrinsic dynamics of the two-stage system [40]. Additionally, this model mathematically validates the effect of HIF-1α on the osteogenic differentiation of hWJ-MSCs. Moreover, based on a multi-stage cell-lineage model and logistic model, our model is sufficiently stable to enable long-term predictions without falling into the trap of population unlimited explosion [41].

By using the existing experimental data, both the cell number and differentiation rate can be predicted with a defined initial cell seeding density and HIF-1α concentration. As such, the optimum incubation time is also obtained. Consequently, we can predict the optimum concentration of HIF-1α and determine the most optimal time for osteogenesis, which is important for efficient tissue engineering. A two-stage cell-lineage model is applicable for predicting the proliferation and differentiation of stem cells, which have two cell lineages. On this basis, the model founded on the initial conditions and existing experimental data can be established to identify the optimum culture conditions in vitro, which will assist in optimizing bone repair in vivo.

Conclusie

In this study, we explored the specific biological effects of HANPs on hWJ-MSCs mediated by HUVECs. Compared to m-HAPs, both np20 and np80 showed slight cytotoxicity in both a time- and dose-dependent fashion. Importantly, the size of the HANPs appeared to have no significant impact on this cytotoxicity. Our data also showed that HANPs, especially np20, were capable of facilitating HUVECs to secrete increased levels of HIF-1α, which directly correlated with the enhanced osteogenic differentiation of hWJ-MSCs via the activation of the ERK1/2 pathway (Fig. 9). More remarkably, the results from the two-stage cell-lineage model suggested that HIF-1α exerted a dose-dependent stimulatory effect on the osteogenic differentiation rate of hWJ-MSCs. Additionally, the optimum concentration of HIF-1α and incubation time were estimated based on the initial conditions using an in vitro model, which could be invaluable in the future for tissue engineering applications. Collectively, these observations provide evidence that HANPs may improve bone regeneration by modulating cell–cell interactions.

A schematic illustration of the possible mechanisms

Beschikbaarheid van gegevens en materialen

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Afkortingen

HA:

Hydroxyapatite

HANPs:

HA nanoparticles

hWJ-MSCs:

Human umbilical cord Wharton’s jelly-derived mesenchymal stem cells

HUVECs:

Human umbilical vein endothelial cell

m-HAP:

Micro-sized HAP particles

PBS:

Fosfaatgebufferde zoutoplossing

ECM:

EC medium

2-MeOE2:

2-Methoxyestradiol

ELISA:

Enzym-linked immunosorbent assay

RUNX-2:

Runt-related transcription factor 2

Col I:

Type I collagen

OCN:

Osteocalcin

ALP:

Alkaline phosphatase

SOX 2:

SRY-related HMG-box 2

ERK:

Extracellular signal-related kinases

VEGF:

Vascular endothelial growth factor