In recent years, scholars have studied whether the pods of beans and other seed shattering are affected by pod length, width, thickness, and degree of curvature. Chiapparino et al. [15] overexpression of GmAGL1 gene in Arabidopsis indicated that there may be a relationship between pod length and fried pod. In leguminous crops, pods with a large degree of curvature and sickle-shaped or long pods are not easy to fry. Krisnawati et al. [16] studied the pod thickness related to soybean frying pod. Murgia et al. [17] and other studies found that the number of broken soybean pods per plant is positively correlated with petal weight, and negatively correlated with traits such as hundred-seed weight, pod perimeter, area, width, and height. Among them, the important factor that regulates seed shattering is the width-to-thickness ratio of the pods. Lou et al. [18] found that the seed shattering rate was negatively correlated with the pod’s width-to-thickness ratio. Tsuchiya et al. [19] and Tiwari et al. [20] study the relationship between size and seed shattering rate found that in soybean varieties resistant to seed shattering, the width to thickness ratio of the pods was generally smaller, but whether a smaller width to thickness ratio would inhibit the occurrence of seed shattering has not been confirmed. In peas, there is no significant difference in the width-to-thickness ratio of seed shattering. Therefore, some scholars believe that the occurrence of seed shattering is not directly related to the width-to-thickness ratio of the pods.

The pods are composed of two single sheets of valves, and the central cavity is where the seeds live. Seed shattering originates from the cracks in the abdominal sutures. Christiansen et al. [21] and Summers et al. [22] analyzed the microstructure of the back suture and abdominal suture cross-section of the pod skin during the mature period of soybeans proposed there is a similar structure similar to the cruciferous dehiscence zone (DZ) at the suture line of the abdomen. As the pod matures, the dehiscence zone gradually evolves into a non-ligninized delamination (SL) about two cell layers thick [23]. Degradation of the delamination and separation of parenchyma can improve seed shattering ability [24,25]. In soybeans, the excessive lignification of the fibrous cap cells connecting the two valves in the ventral suture layer of the pod increases the bonding strength of adjacent cells, thereby forming the characteristics of seed shattering of cultivated soybeans. The study of Dong et al. [26] suggested that the thickening of the secondary cell wall of FCC cells is the main factor for the resistance of cultivated soybean varieties to seed shattering. For the degradation of the intercellular layer in the dehiscence zone, the differentiation of the pod wall is essential. The pod wall can be divided into three functional layers, namely exocarp, mesocarp and endocarp [27]. In the process of seed shattering, due to the reduction of cell wall adhesion in the cracking zone, and the tension generated by the mechanical properties of the endocarp and exocarp cells of the pod (the tension produced during the shrinkage process caused by the difference between the endocarp and the exocarp [28]) causing the pods to crack. The key factor that produces the required tension is the geometric arrangement of the cell wall and its histological characteristics (cell wall thickness and composition, lignification, hydration). Suanum et al. [29] found that the composition and characteristics of the cell wall of the pod are related to the fragmentation of cowpea and wild cowpea. In the late stage of maturity, once the valve is fully lignified and the valve attachment becomes weak, the increase in the size of the vacuolated cells in the sidewall of the cracking line will promote the cracking. In other plant species, due to dehydration, the seed shattering is also related to the difference in the expansion of the inner thick-walled cell layer [30]. Murgia et al. [31] proposed that the lignification of the inner valve layer (LFL) contributes to seed shattering, while the study of Bruce et al. [32] showed that the absence of LFL or the presence of a low level of lignification would reduce the stability of the valve, thereby reducing the stability of the valve during drying. In addition, the fracture structure of the pod wall was found to be correlated with the helical moderately, and has visual cracking in some environments.

In summary, the thickness of the secondary cell wall of the lignified FCC cells and the FCC cells in the ventral suture, the structure of the pod wall, and the cause of tension are the main factors that regulate the cracking of the pod. It is believed that subsequent studies will be more detailed analysis for the cytological structure characteristics of soybean pods and some unknown structures and their effects.

Seed shattering is closely related to its own characteristics, and genetic characteristics are the main factors affecting seed shattering [33]. Research by Li et al. [34–36] showed that different soybean varieties have significant differences in the resistance to seed shattering under the same environmental factors, and there is a certain correlation between the differences in pod structure and water content between different soybean varieties. Nilmani et al. [37] studied the seed shattering rate of different soybean varieties 10 days after the maturity period, and the seed shattering rate was between 0.67% and 67.05%, implying that genetic background has a great influence on soybean seed shattering. For this reason, researchers have studied the genetic mechanism of seed shattering on many model crops such as soybeans, peas and lentils. These studies have shown that the pathways and loci that control seed shattering is widely conserved in species [38], A large number of studies have shown that soybean seed shattering is not determined by a single gene, but a quantitative trait co-regulated by a major gene and multiple minor genes [39].

For the initial research of soybean seed shattering, the main focus was on the impact of environmental factors on soybean seed shattering and corresponding measures were proposed to reduce seed shattering so as to avoid the yield loss caused by seed shattering. In 1957, Metcalfe et al. [40] found that mature Lotus japonicus pods showed two phenotypes of cracked pods and non-cracked pods at an environmental humidity of 29.5% and 35%. Relative humidity was first proposed as the main environmental factor affecting legume crop blast pods. Studies have shown that the frying rate of French croissant pods at low temperature and high humidity is significantly lower than that at high temperature and low humidity [41]. Subsequently, Cavines et al., successively proposed that relative humidity is an important environmental factor affecting soybean seed shattering [42,43]; in addition, Zhang et al. [34], not only verified the above views, but also put forward that there is no obvious correlation between the seed shattering rate and the length of sunshine time and the average daily temperature. Yang researched that seed shattering occurred when the moisture content of beans and seed shattering were lower than 25% and 15%, respectively, indicating that the water content of soybeans is also closely related to seed shattering and also proposed that the spatial distribution of pods in identical soybean does not significantly link [44]. In addition, Tsuchiya et al. found that sudden temperature changes can induce soybean pods, which is of great significance for actual production [45]. Travis found that under drought conditions, the twisting force of the pod wall or the structural strength of the cracked area in the pod suture will change, resulting in soybean seed shattering [46].

To sum up, the relative humidity in the air is a direct factor that affects soybean seed shattering, and the rapidly changing environmental factors are the key factors leading to the decrease of the moisture content of the soybean pods. A series of environmental factors such as temperature, light intensity, sunshine hours, wind speed, air pressure, and soil moisture indirectly affect soybean pods by changing air humidity. In order to change the impact of the above reasons on seed shattering, we can harvest them timely, and properly spray water or desiccant to reduce the loss caused by seed shattering, but this will affect the cost and improper operation will cause problems such as mildew of the pods.

Soybean harvesting is a key link in the soybean production process. The use of appropriate harvesting machinery to complete the soybean harvesting operations in a timely manner is an important guarantee for high soybean yield. The mechanical harvest loss of soybeans has always been a serious problem in mechanized soybean harvesting. Langowski et al. [47] conducted a long-term experimental study on the mechanical harvest loss of soybeans. The total mechanical harvest loss rate was 9.8% to 19.3%, and the header loss accounted for 80%, of which shattering loss accounted for 55%. The test results of Deng Si et al. showed that branching and shattering accounted for 94% of the loss. Pods are easy to burst due to the biological characteristics of the crop. Li et al. [48] studied the biological and mechanical characteristics of soybean varieties in the Huanghuaihai area, and provided some parameters for the header and threshing system of the Huanghuaihai soybean combine harvester. Many domestic scholars have made statistics on the start time of soybean frying and found that most varieties start seed shattering 6–12 days after soybeans are fully mature. So timely harvest is also necessary to avoid large soybean production caused by seed shattering. In summary, timely harvesting and modification of machinery based on the biological characteristics of soybeans can avoid seed shattering and reduce yield loss during harvesting.

Physiologically, the crush resistance of the pod is essentially related to the amount and composition of carbohydrates in the pod. Carbohydrates, such as cellulose and lignin, are important components of the cell wall of the pod and directly affect the resistance of the pod to mechanical breakage. Cellulose guarantees the toughness of the cell wall, while lignin is an important phenolic compound with a complex structure in plants, which enhances the hydrophobicity and hardness, physical strength and water conductivity of the cell wall. Cellulase can also degrade the cell wall. Agrawal et al. [50] measured the cellulase activity in soybean pods and found that the cellulase activity in the pod germplasm at the mature stage was significantly increased and the high content of cellulose (and hemicellulose) and lignin in the separation layer will hinder the separation of cells in the separation layer and reduce pod damage. Kappara S showed that the seed shattering determined by scoring the seed shattering of cowpea/cowpea has the highest correlation with the pod hemicellulose content, but not the correlation with the pod lignin content [5].

In addition, in the process of seed shattering, β-glucase and polygalacturonic acid endonuclease are the key enzymes that play a role. The activity of glucose hydrolase in the dehiscence zone of the easy-explosive pod germplasm of Brassica is increased and the cell wall is degraded. For the conversion and distribution of β-glucase, this enzyme may be involved in the entire process of layer separation [51]. The endogenous polygalacturonic acid is involved in the degradation of the intercellular layer and is related to the cell separation during the pod shedding process. The pectin is related to the adhesion between the cells. Pectin degradation is found in the plant germplasm that is easy to shatter. Severely, in the process of crop maturation and senescence, increasing the activity of glucose hydrolase and endogenous polygalacturonase can significantly reduce the content of pectin in the cracking zone.

As mentioned above, these enzymes participate in the seed shattering process by separating the cell layer and degrading the cell wall, as well as affecting the adhesion between the cells. However, the activity of these enzymes in the pod, how to express, and the mechanism of action are still unclear, and further research is needed.

Phytohormones are trace organic substances produced by plants’ own metabolism, also known as plant endogenous hormones, which can regulate plant physiological responses by affecting the activity of enzymes. Among the five endogenous hormones, two are related to β-glucose hydrolase, which can degrade pod cell walls. When the pod is shatter, the content of ethylene in the kernel suddenly increases and a short-term peak appears. At this time, the activity of β-glucose hydrolase increases, indicating that the change of ethylene content indirectly regulates the cracking of the pod. The peak of ethylene content usually occurs earlier than the cracking of the pod suggesting the peak ethylene content may be a signal of pod splitting. Relevant studies have shown that auxin can inhibit the accumulation of β-l, 4-glucohydrolase mRNA in the dehiscence zone to regulate pod dehiscence [52]. Gibberellin plays a role in the establishment of the cell characteristics of the separation layer, but the topical application of cytokinin during fruit development can restore the formation of valve edges and further increase pod dehiscence. Tu [53] found that the synergistic effect of auxin and gibberellin in the cracking zone can increase the activity of vegetable soybean endoglucanase and pectinase, accelerate the decomposition of cellulose and pectin, and form the characteristics of easy-fried pods.

In summary, these findings indicate that several plant hormones are involved in the differentiation of DZ and that their biosynthesis and response are essential. Although the role of hormones in the development of DZ has been extensively investigated, there are relatively few studies on how these hormones coordinate in DZ. Therefore, in the future, it is necessary to in-depth study of the molecular mechanism and interaction of plant hormones behind the DZ differentiation of cracked pods.

In the past 20 years, some QTLs and genes that control pod dehiscence have been identified, revealing that a major QTL and several minor QTLs co-regulate pod dehiscence. Bailey et al. [59] first used 140 restriction fragment length polymorphism (RFLP) markers to identify the main QTL. Funatsuki et al. [55] detected this major QTL and named it qPDH1. Furthermore, studies have shown that qPDH1 is the main QTL related to pod dehiscence. Funatsuki et al. [60] cloned the gene pdh1 near the QTL “qPDH1” reported on chromosome 16, and promoted pod dehiscence by increasing the twisting of the dry pod wall. In addition, Dong et al. [61] analyzed the changes caused by domestication between cultivated soybeans and wild soybeans and identified the NAC gene SHAT1-5 (Glyma16g02200), proving that this gene activates secondary cell wall biosynthesis and promotes the thickening of fibrous cap cells in pod sutures.

Recently, some new QTLs have been discovered on chromosomes 1, 4, 5, 6, 8, 9, 11, 17, 18, and 20 through genome-wide association analysis (GWAS). 16 QTLs were identified from the high-density linkage maps of two RILs, including the qPDH1 locus, and several new QTLs were identified on chromosomes 2, 6, 11, 13, 14, 16, and 20 [62]. At present, the mechanism of bean pod shatter prevention is not yet fully understood, and the genetic backgrounds of the mapped parents are different, and the genetic effects explained by the mapped QTLs are also different. Therefore, other QTL information from different genetic backgrounds is essential for the development of shatter-resistant soybean varieties.

So far, there have been few studies on the cloning and function of the fried pod gene in soybeans. In 2014, Funatsuki et al. [60] obtained the fried pod-resistant gene PDH1 through map-based cloning. Because this gene contains an early stop codon which leads to loss of function so it can reduce the twisting force of the pod skin to reduce the soy pod frying ability. In the same year, Dong et al. [61] found that the 116 kb region of chromosome 16 is related to fried pods, and successfully cloned a seed shattering gene in this region, named SHAT1-5, and proved that SHAT1-5 gene regulates FCC cells, resulting in the thickening of the secondary cell wall and the formation of anti-explosive pods. In 2017, Qing et al. [63] performed a homology comparison of the Arabidopsis FUL gene to obtain the soybean fried pod gene GmAGL8, and found that overexpression of this gene can reduce the rate of soybean and Arabidopsis seed shattering. There are also cloning and functional studies of seed shattering genes in other plants and the most studied is Arabidopsis thaliana. Details are shown in Table 2. Dinneny’s research on the genetic mechanism of Arabidopsis silique seed shattering is shown in the Fig. 1. In summary, seed shattering is a complex process of multi-gene regulation, and more genes related to seed shattering need to be discovered.