Manual database search yielded 586 articles (Figure 1). After the first screening stage of titles and abstracts, 84 articles were selected for full-text review. After the full-text screening, 68 articles were excluded, leaving 16 articles that met the inclusion criteria. They were excluded because they failed to assess papers, were not animal models (e.g., human sample), did not include interventions (e.g., observation of normal tendon development over time and gene modification model of the tendon), and did not have any in vivo model (e.g., in vitro and ex vivo), were not related to natural development (e.g., healing, engineering research), did not match the age (within four weeks from the day of birth), and the outcome did not include tendons (e.g., only muscle-tendon junction or enthesis) (Figure 1).
Experimental animal models were categorized into four species
The intervention methods and tendon development process vary with animal species, so we categorized by animal species. Table 1 shows the characteristics of the experimental model in the included studies. Four different species were used; chicks (n=13), mice (n=5), zebrafish (n=3), and rats (n=2). The age of the chick models was between stage (st)18-19; it was converted into the embryonic day E2-E3 (2 days from intervention at stage 4 and 5 (Hamburger and Hamilton, 1951)). Mice aged E12.5 to P0 and rats aged P0 to P10 were used as models. Zebrafish from 48 h post-fertilization (hpf) to 98 hpf were used as models. Most studies used embryos, and only three studies used postnatal animals.
This scoping review identified interventions that decrease and increase the mechanical force to understand the role of mechanobiology in tendon development. The interventions that reduced mechanical force involved 14 approaches; in contrast, only three ways were used to increase mechanical force. The intervention type for reducing mechanical force was categorized into five groups for various animal intervention methods: transgenic, surgery, drug treatment, locomotion, and added stimulation (Table 2). Five studies used multiple models. Three studies combined interventions that changed the mechanical force and techniques that inhibited specific signaling. In situ hybridization (section or whole mount) was frequently used to analyze tendons. Electronic microscopy, immunohistochemistry, transmission electron microscopy, and mechanical testing were used for assessing the tendons. The other methods are shown in Table S1. Some biological markers tested the effect of mechanical force. The most frequent marker was Scx, which is common in many studies. Scx mutant mice showed a dramatic defect in tendon differentiation. Scx is recognized as required for tendon differentiation and formation(Murchison et al., 2007). So, downregulated Scx expression in the intervention model means inhibiting tendon development. GFP fluorescence, in situ hybridization (ISH), and polymerase chain reaction (PCR) can also detect Scx expression.
The lack of consensus regarding standard intervention methods
We categorized types of intervention because that improved to catch the sight of defining mechanical force. Types of intervention were identified as ones that decreased or increased the mechanical force. The intervention type of decreased mechanical force was divided into four groups. The transgenic group included several muscle-deficient factor mutant models that included Myf5–/– Myod1–/– double mutants (Brent et al., 2005; J. W. Chen and J. L. Galloway, 2014), Pax3 (splotch or splotch delayed) mutants (Schweitzer et al., 2001), muscular dysgenesis (mdg) mutant(Huang et al., 2015), and voltage-dependent L-type calcium channel subtype bata-1 (cacnb1) mutant (A. Subramanian et al., 2018). Full-length mRNA encoding codon-optimized a-bungarotoxin (aBtx) was injected into the muscle (A. Subramanian et al., 2018). Most studies used the mouse model, except for the cacnb1 mutant and aBtx mRNA intervention, which were in zebrafish. Myf5–/– Myod1–/– double mutants do not contain any muscle progenitors, resulting in a muscle-less limb (Kassar-Duchossoy et al., 2004). Pax3 is required for the establishment of muscle progenitor cells in the limb, and mutations in Pax3 (Splotch) and Pax3 (Splotch-delayed) cause a severe defect in limb muscle formation (E Bober et al., 1994). mdg is an autosomal recessive lethal mutation that results in the contraction of skeletal muscles (B A Adams and Beam, 1990). Full-length mRNA encoding codon-optimized aBtx was injected in zebrafish to prevent skeletal muscle contractions (Swinburne et al., 2015). The limb was paralyzed in the cacnb1 homozygous mutant (Zhou et al., 2006).
Surgical procedures can remove muscle or neural tube in a region-specific manner. The surgery group experimented with coelomic wing graft (F. Edom-Vovard et al., 2002; Kardon, 1998), neural tube ablation (F. Edom-Vovard et al., 2002), and dermomyotome removal (Ava E Brent et al., 2003) in chicks. Coelomic wing graft was performed in the lateral plate areas corresponding to the future wing buds, or migratory myogenic cells were isolated from chick embryos. This weakened the wing muscle. Neural tube ablation was performed on embryos before the exit of ventral root fibers to produce complete aneural wings. Dermomyotome removal is a method of surgically removing the dermomyotome before myotome formation, making it muscle-less.
In addition, interventions using four types of drugs were also performed. They were decamethonium bromide (DMB) (Havis et al., 2016; J A Germiller et al., 1998; X S Pan et al., 2018), pancuronium bromide (PB) (Havis et al., 2016; X S Pan et al., 2018) , d-tubocurarine (C Beckham et al.), and botulinum toxin (W.G.Hopkins, 1984) . Botulinum toxin was used in mice, while the other interventions were used for chicks. DMB prevented the effects of acetylcholine at the neuromuscular junction and depolarized the end-plate region, inducing rigid paralysis (Paton and Zaimis, 1951). PB blocked the response to acetylcholine and is characteristic of blocking drugs of the non-depolarizing type. Thus, chicks treated with doses of PB were paralyzed and flaccid (W R Buckett et al., 1968). D-tubocurarine is an acetylcholine receptor antagonist that inhibits neuromuscular activity (L T Landmesser and Szente, 1986; Paton and Zaimis, 1949). Botulinum toxin, an acetylcholine receptor antagonist, decreased the amount of acetylcholine interacting with receptors, thereby reducing muscle contraction and motility (Pittman and Oppenheim, 1978); PB, d-tubocurarine, and botulinum toxin induced flaccid paralysis.
In contrast, mechanical force was increased by injecting a group of drugs, which included only 4-aminopyridine (4-AP) (X S Pan et al.), a neuromuscular blocking drug that blocks the potassium channels in neurons. When applied to chicks, it stimulated the release of the neurotransmitter, acetylcholine (ACh) and enhanced its availability at the synaptic cleft (Heywood et al., 2005). Therefore, the 4-AP model induces high-frequency movement and hypermobility (X S Pan et al.).
The locomotion group focused on rodent locomotion (spinal cord (S. K.Theodossiou et al., 2021)) in the postnatal phase of rats. The developing locomotor behavior during the postnatal period was believed to increase mechanical loading for limbs. Rats with spinal cord transection did not show complete weight-bearing locomotion; hence, they were used as model systems to disrupt locomotor development in neonates. The postnatal rats showed changes in spontaneous posture and locomotion during the early postnatal week (H. E. Swann and M. R. Brumley, 2019). This locomotion development affects the limb mechanical loading (S. K. Theodossiou et al., 2019).
The last group includes the electrically stimulated model (A. Subramanian et al.) that induces muscle contractions in zebrafish. This intervention was applied to both normal and paralyzed muscles to increase mechanical force.
Lack of mechanical force inhibited tendon maturation
We summarized the model to investigate the role of mechanical force in tendon development throughout Scx expression. Tendon development of individual models is described below and summarized in Table 3 and Table 4. The most standard method to control the mechanical force of the tendon is to arrange the force depending on the muscles. Many researchers reported the way to modulate muscle function using a variable animal model. The defined mechanical force is roughly divided into three types. One of the methods is the Muscle-less model using a surgical extraction, Pax3 mutant mice model. The others are muscle paralysis models that operate muscle contraction, such as DMD injection. The third is weight-bearing. This section showed how each type of mechanical force affects the expression of Scx.
At first, we show the relationship between muscle-less and Scx expression. Myod1 Myf5-deficient zebrafish expressed Scx 53-58 hpf, but not at 72 hpf (J. W. Chen and J. L. Galloway, 2014). Myf5–/–; Myod1–/– double mutant muscle-less mouse embryos survived Scx expression in the limbs but not in the epaxial region at E13.5 (Brent et al., 2005). In Pax3 mutant embryos, which were muscle-less similar to the Myf5–/– Myod1–/– double mutant, Scx expression in limbs was not affected at E12.5 (R Schweitzer et al., 2001). In the E16.5 Pax3 mutant, Scx-GFP was not detected in the zeugopod tendon. However, it persisted in the autopod tendon segments at E18.5 (Huang et al., 2015). At E16.5, Scx GFP was observed in the zeugopod of the mdg mutant. Moreover, muscle-lessness can also be induced by surgical manipulation. Induction of Scx was not observed after surgically removing the dermomyotome before myotome formation (Ava E Brent et al., 2003). Coelomic wing graft surgery indicated that tenascin reduced the proximal and intermediate tendon but rescued the distal limb (Kardon, 1998). Another study on coelomic wing graft surgery involving chick embryos at E2 showed Scx expression at the forearm at E6, but not at E10 (F. Edom-Vovard et al., 2002).
The tested model the effect of muscle contraction, in chick E8 and E9, the neural tube ablated embryos showed downregulation of tenascin and Scx in tendons. Although Scx expression in the forearm at E10 (or E9) was limited in both models, the digit expressed Scx at E10 (F. Edom-Vovard et al., 2002). The expression of factors related to tendon development, such as Scx and the transforming growth factor-beta-2 (Tgf-β2), decreased at E6.5 in chicks paralyzed rigidly using DMB, while Tenomodulin (Tnmd) expression decreased at E7.5 (Havis et al., 2016). Through E18, the tendons became uniformly smaller and correlated with reduced chick movement due to paralysis (J A Germiller et al., 1998). In chicks with flaccid paralysis induced by PB, Scx expression decreased at E6.5 (Havis et al., 2016) and LOX activity was reduced at HH43 (E17) (X S Pan et al., 2018). The length of the muscles reduced in botulinum toxin-injected mouse, and the tendons were longer than the muscles (W.G.Hopkins, 1984). The fibrocartilaginous area and elastic vinculum were not formed in chick injected with d-tubocurarine. However, the tendon cells and collagen in immature elastic tendon fibers did not change (C Beckham et al., 1977). The increased mechanical force in the 4-AP model increased the elastic modulus. Zebrafish stimulated by electronic stimuli to restore mechanical force reduced by aBtx-injection showed increase in axial tenocyte projection length compared to that observed in aBtx-injected only animals. Several factors, such as Thrombospondin 4b (Tsp4b), TGFb-induced protein (Tgfbip), and connective tissue growth factor a (Ctgfa2), were upregulated to control levels by electronic stimulation compared to that observed in aBtx-injected animals(A. Subramanian et al., 2018).
The last is weight-bearing model. Theodossiou et al. reported the structural properties and cross-sectional area of the weight-bearing Achilles tendon at P10 were higher than those of the non-weight-bearing phases P1 and P5 in the rat model (S. K. Theodossiou et al., 2019). But there is no research measuring Scx expression using the weight-bearing model.
Intervention for detecting molecular mechanism
In the above section, we focused on Scx expression. Moreover, we summarize several markers regulated by mechanical force along with Scx. One study confirmed the relationship between expression and molecular mechanism by reimplanting a source of fibroblast growth factor 4 (Fgf4) in the aneural chick limbs and muscle-less chick wing (Havis et al., 2016). Fgf4 was not expressed in aneural and muscle-less wings; hence, reimplanting was performed to analyze the possible effects of Fgf4 removal from aneural muscles on tendon markers. Scx and tenascin were downregulated in the aneural limbs and muscle-less wings. Consistent with this, both models showed that grafts of Fgf4 cells rescued the expression of the tendon-associated molecules, Scx and tenascin. Another study tested whether Fgf rescued tendon gene expression without muscle contraction in DMD-injected chick. mFgf4/ RCAS-producing cells were grafted into forelimb buds. While immobilization following DMD induced a drastic decrease in Scx, Tnmd, and Thrombospondin 2 (Thbs2) expression, the mFgf4-paralyzed-limbs significantly upregulated Scx, ETS translocation variant 4 (Etv4) also known as polyoma enhancer activator 3 (Pea3), and sprout 2 (Spry2). The relative mRNA levels of Tnmd, Tgf2, and Tgfβ3 did not change under this condition. These results suggested that the downregulation of several genes may be a molecular effect induced by the muscle-less state or inhibition of contraction. The muscle may require both mechanical and molecular aspects.