1. Cell cultures

In this project, we worked with the continuous line of murine long bone MLO-Y4 osteocytes Cx43 +HPP+, as a control, and deficient in Cx43 (Cx43 -HPP-), which were ki1. Cell culturesndly provided by Dr. L. I Plotkin.

The cells were cultured at a concentration of 24,000 cellsHPPcm², with ɑ-Modified Eagle's Medium (ɑ-MEM) (Gibco^TM, Thermo Fisher Scientfic, Spain), supplemented with 2.5% calf serum (Calf Serum; CS), 2.5% fetal bovine serum (FBS), 1% L-Glutamine, 1% PenicillinHPPStreptomycin and Puromycin (from Streptomyces alboniger, Sigma Aldrich, BioReagent, Merck, Spain) at a concentration of 10 μgHPPml.

To promote the development of the primary cilium, the cells were grown in depletion medium, composed of ɑ-MEM (Gibco^TM) supplemented with 1% penicillinHPPstreptomycin and puromycin at a concentration of 10 μgHPPml, for 24 hours.

All surfaces on which these cells were cultured had to be previously collagenized, using for this matter type I collagen at 0.01% acetic acid, to simulate the collagen matrix where the osteocytes are embedded in vivo. Cells were kept at 37ºC and 5% CO₂.

2. Western blot

The cellular total protein was extracted using RIPA buffer (Sigma-Aldrich, Merck, Spain), supplemented with protease and phosphatase inhibitors (Calbiochem, Merck, Spain).

Subsequently, these proteins were quantified using bicinchoninic acid (BCA) (PierceTM BCA Protein Assay Kit, Thermo Fisher Scientfic, Spain), which generates a colorimetric reaction detectable at 562 nanometres (nm). To carry out the reading, the Varioskan Flash plate reader (Thermo Fisher Scientfic) was used, with SkanIt Software 2.4.3 RE.

Protein extracts (20 µg) were separated by means of a 10% polyacrylamide gel under reduced conditions. They were later transferred to a nitrocellulose membrane (Bio-Rad, Hercules, California, USA). The blocking was carried out with 5% milk powder, in Tris-saline buffer with 0.05% Tween20 (TTBS), for 1 hour under stirring at room temperature. The following primary antibodies were then stirred between 15-18 hours at 4ºC: anti-Phosphop-44HPP42 MAPK (Erk1HPP2) (Cell Signaling, Beverly, Massachusetts, USA), anti-p44HPP42 MAPK (Erk1HPP2), anti-Cx43 (Sigma Aldrich, ST. Louis, Missouri, USA) and anti-tubulin (Sigma Aldrich). These are all polyclonal antibodies produced in rabbits, except for anti-tubulin which is a monoclonal antibody produced in mice. It was then incubated for one hour at room temperature, with the corresponding IgG coupled to peroxidase, and the membrane was developed by chemiluminescence with Clarity^TM Western ECL substrate (Bio-Rad, Life Science Research, Spain). The intensity of the bands was quantified by densitometry, using the DNR Bio Imaging System MF ChemiBIS3.2 and Gelcapture and QuantityOne^TM (Bio-Rad) programs.

3. Immunofluorescence

30,000 cellsHPPwell were cultured from the multiwell plates (Falcon^®, Thermo Fisher Scientfic, Spain). The cells were grown until they reached 80% confluence; and subsequently depletion medium was added for 24 hours to induce the formation of the primary cilia. The cells were then fixed with 4% paraformaldehyde (PFA) and permeabilized with Triton X-100 at 0.5%. Next, the blocking solution, composed of bovine serum albumin (BSA) at 10%, supplemented with goat serum at 5%, was added for 1 hour. After that, the following primary antibodies were stirred for 15-18 hours at 4ºC: polyclonal anti-Cx43 produced in rabbits (Sigma Aldrich) and monoclonal anti-ɑ acetylated tubulin produced in mice (Sigma Aldrich), to observe the primary cilium. The secondary antibodies were then arranged: Alexa fluor^® 488 goat anti-mouse cilium (Invitrogen Molecular probes, Thermo Fisher ScientificTM, Spain), and Cx43, Alexa fluor^® 568 anti-rabbit IgG (Life technologies, Thermo Scientific^TM, Spain) (1: 1000 dilution in BSA at 10% and goat serum at 5%). After 1 h of incubation, 4'-6-diamidino2-phenylindole (DAPI) was added. Nuclei, the primary cilia and Cx43 were seen on the fluorescence microscope (Leica CTR 6000). The merging of the individual images of the primary cilium, Cx43 and cell nuclei into one single image was performed using the ImageJ program.

4. Mechanical stimulation by fluid flow (FF) and by PTHrP

To perform stimulation by FF and PTHrP, cells were cultured on teflon-bound glass slides at a density of 25,000 cellsHPPcm². The FF technique is based on the constant bombing of culture medium over the cell monolayer using a peristaltic pump (Flexcell International Corp., Hillsborough, North Carolina, USA.) at 10 dynesHPPcm² in a hermetically closed circuit for 10 minutes. The time and frequency settings were established using the Master Flex Peristaltic Pump 2010 program. On the other hand, stimulation was performed with PTHrP (1-36) (Bachem, Bubendorf, Switzerland) at a concentration of 10^-7 molar (M), for 10 min. The same number of cultures, but with cells that were not subjected to any stimuli, served as static controls (SC).

5. Statistical analysis

In the statistical analysis of the results, data are expressed as mean ± standard deviation of at least two experiments carried out in triplicate. It was performed using the GraphPad Prism 8 program (GraphPad software, La Jolla, California, USA). To compare means between two groups, the nonparametric Mann-Whitney test was carried out, and to compare means of more than two groups, the non-parametric Kruskal-Wallis test was used. For mulptiple pairwise comparison we used Dunn’s test. The established confidence interval was 95% in all statistical tests. So the results with a value of p<0.05 were considered statistically significant.

1. Effect of Cx43 on the development of the primary cilium in Cx43 +HPP+ and Cx43 -HPP- cells

The results obtained by Western blot indicate that in Cx43-deficient cells the expression of this protein decreases significantly in comparison with control cells Cx43 +HPP+ (W=0, p=0.029), which allows us to confirm the deficiency of this protein (Figure 1). In addition, the presence of the primary cilium was characterized by the expression analysis of the IFT88 protein. IFT88 was used as a marker for the primary cilium presence, since it is a highly abundant protein in this organelle, as it is involved in intraflagellar transport, needed for ciliogenesis¹⁸. We observed that the IFT88 protein expresses in a similar way, regardless of the deficiency in Cx43 (W=4, p=0.343).

Figure 1.  Expression of Cx43 and IFT88. A) Result of the nitrocellulose membrane development for the analysis of Cx43, IFT88 and tubulin as loading control; total cellular protein extracts were used (25 µg). The molecular weight (MW) marker was placed in the first lane; the next four lanes belong to four replicas of control MLO-Y4 cells (Cx43 +HPP+) and the last four correspond to four replicas of Cx43-deficient MLO-Y4 cells (Cx43 -HPP-). B) Mean ± standard deviation of Cx43HPPtubulin. C) Mean ± standard deviation of IFT88HPPtubulin. Protein levels normalized in contrast to tubulin *p<0.05 

Likewise, Cx43 and primary cilia were analyzed by immunofluorescence to try to determine the possible interaction between these two mechanosensors. Figure 2 shows both Cx43 +HPP+ and Cx43 -HPP- cells develop a primary cilium, which is evidenced by the presence of the primary anti-α-acetylated tubulin antibody, which originates from the cell surface. Also, it can be seen that Cx43 appears mainly in the cell membrane, possibly forming GJs. Furthermore, it appears that the primary cilium does not co-localize with Cx43. On the other hand, we performed counts of the cells with primary cilium from nine photographs taken in different fields using the fluorescence microscope. In each image the number (No.) of total cells (cel.) and the number of cells with a presence of primary cilium were quantified, and we calculated the ratio (No. of cells with ciliumHPPNo. of total cells), both for the Cx43 +HPP+ and for Cx43 -HPP- cell lines (Table 1 and Figure 3). In these results, we found that the number of cilia formed did not significantly differ (p=0.089) between the two cell lines (Cx43 +HPP+ and Cx43 -HPP-).

Figure 2.  Immunofluorescence performed with osteocytes of the MLO-Y4. A) Cx43 +HPP+ and B) Cx43 -HPP- line. Cell nuclei were stained with DAPI (blue), anti ɑ-acetyl-tubulin antibodies were used to mark the primary cilium (green), anti-Cx43 antibody to mark Cx43 (red) and merged. The images were taken with an immersion objective at 63X. Scale bar = 16 μm 

Table 1.  Counts of cells (cel.) with presence of primary cilium in the MLO-Y4 Cx43 +HPP+ and Cx43 lines 

Figure 3.  No. of cells with ciliumHPPNo. of total cells (%) in the Cx43 +HPP+ and Cx43 -HPP- lines. Results are expressed as mean ± standard deviation 

2. Effect of mechanical stimulation and PTHrP on Cx43 +HPP+ and Cx43 -HPP- cells

In several studies it has been determined that mechanical stimulation inhibits osteocyte apoptosis, through a mechanism that involves phosphorylation of MAPKs as ERKs¹⁶. The aim of this experiment was to analyze whether Cx43 deficiency altered the effect of mechanical stimulation on osteocytes, and to determine whether Cx43 is involved in the activation of the PTH1R receptor, after stimulated by one of its PTHrP ligands. The effects triggering mechanical stimulation and PTHrP stimulation in MLO-Y4 cells, the P-ERK expression was analyzed using Western blot.

The Kruskal-Wallis test was applied to establish the differences between the Cx43 +HPP+ and Cx43 -HPP- groups for the conditions: SC, FF and PTHrP. The test result pointed at differences between some of the groups (p=0.0005). To determine between which groups, Dunn’s Multiple Comparison Test was performed, showing that both mechanical stimulation (FF) (p=0.049) and PTHrP stimulation (p=0.017) induce a significant increase in ERK phosphorylation in contrast to non-stimulated cells (static control, SC); in the MLO-Y4 Cx43 +HPP+ cell line. This result shows that extracellular stimuli favour the activation of the SrcHPPERK signaling pathway, which promotes cell survival. In the case of MLO-Y4 Cx43 -HPP- we observed that P-ERK does not increase significantly in contrast to CE (p=0.955) after mechanical stimulation (FF). However, when stimulated with PTHrP, ERK phosphorylation increases significantly (p=0.025) (Figure 4).

Figure 4.  P-ERK analysis after mechanical stimulation and with PTHrP. After mechanical stimulation and stimulation with PTHrP, changes in protein levels were evaluated by Western blot, and those regarding P-ERK in total cell protein extracts, cell lines Cx43 +HPP+ and Cx43 -HPP-, stimulated or not (static control, SC). Total ERK and tubulin were used in order to normalize. Results are expressed as mean ± standard deviation of a duplicate experiment of each condition vs. SC