Comparison of root dry matter content and root dry matter weight in sweet potato genotypes cultivated by transplanting and direct planting

ABSTRACT Direct planting (i.e. the planting of seed roots) can reduce the labor requirements of sweet potato (Ipomoea batatas (L.) Lam.) cultivation. To broaden the breeding material for high dry matter productivity in direct planting, root dry matter content (DMC) and root dry matter weight per hill (DMW) were compared under both direct planting and conventional transplanting by using 28 sweet potato genotypes cultivated in 2020 and 2021. Unlike in the case of the mother root DMC, a significant positive correlation between the daughter root DMC in direct planting and the root DMC in transplanting was observed over two years (p < 0.001). Additionally, a significant positive correlation was observed between the total root DMW in direct planting and root DMW in transplanting (p < 0.001). These results suggest that the dry matter productivity of genotypes in direct planting cultivation can be presumed from the performance of the transplanting cultivation. GRAPHICAL ABSTRACT


Introduction
Sweet potato is generally propagated using stem cutting, termed 'transplanting cultivation'.The preparation for stem cuttings and transplanting requires more than one third of the total labor associated with sweet potato cultivation in Japan (MAFF, 2020).As an alternative planting system, 'direct planting cultivation', in which small storage roots as we call 'seed roots' are planted instead of stem cuttings, has been studied since the 1940s in Japan and the USA reviewed by George et al. (2011).This is similar to the propagation system of potato (Solanum tuberosum L.), which is already mechanized.For this reason, direct CONTACT Takeo Sakaigaichi gaichi@affrc.go.jp planting of sweet potato can reduce labor requirements (George et al., 2011;Shikata, 1976).Ikemoto (1971) calculated the working hours for direct planting, and proved that less labor were required compared with the conventional transplanting.Regarding economic analysis, Kubota et al. (2000) simulated that the production cost by the direct planting became lower than that by the transplanting as the productive scale extended.This type of planting is not used in large-scale anywhere in the world in sweet potato because most genotypes produce enlarged 'mother roots' (George et al., 2011;Kobayashi, 1968).In direct planting in sweet potato cultivation, two types of roots can be obtained at harvest.One is 'mother roots', which develop from the planted seed roots, and the other is 'daughter roots', which are newly formed after planting.Harvested mother roots are generally enlarged in most of genetic resources, presenting a deformed shape and an inferior quality compared to daughter roots (Adachi et al., 2012;Shikata, 1976).For these reasons, mother roots have no commercial values.However, long-term breeding efforts to generate cultivars for direct planting (Kobayashi, 1968;Kusuhara et al., 1972) have improved breeding materials with regard to resistance against mother root enlargement.As a result of previous studies, several promising cultivars with enough 'daughter roots' have been released (e.g.'Tamaakane', Sakai et al., 2011).
Sweet potato is used as raw materials for alcohol production and so on.Therefore, starch yield, as well as daughter root yield are taken into consideration when screening clones in the breeding program of direct planting.As demonstrated by previous reports (George et al., 2015;Shigemura et al., 1931;Tumwegamire et al., 2011), it is well known that root starch content (%, fresh matter) has a positive correlation with root dry matter content (DMC, %, fresh matter).Therefore, DMC is commonly used as a simple indicator of starch content, particularly in the early stages of breeding programs.
Except for 'Suzukogane' (Sakai, 2017), the current direct planting cultivars such as 'Tamaakane' and most of the breeding lines in Japan generally have lower DMC than the leading cultivars such as 'Koganesengan' and 'Shiroyutaka'.An effective way to improve DMC and dry matter yield of breeding lines for direct planting is to cross elite genotypes with high DMC and dry matter yield, which were already screened in the conventional breeding program.However, the correlation of the DMC and the dry matter yield between the sweet potato genotypes grown by transplanting and direct planting was rarely examined.Kobayashi (1972) studied this topic and compared root starch content and root yield of sweet potato grown by transplanting and direct planting; it indicated that these traits in direct planting could be presumed from the performance in transplanting because of the significant correlations of them between different cultivations.Nonetheless, these correlations have not been studied on the genetic resources, including recent cultivars and breeding lines bred for direct planting.Furthermore, Kobayashi (1972) examined the correlations only for asingle year, and a multi-year evaluation is required to discuss this correlation because a significant genotype × environment interaction and yield instability exists in direct-planted sweet potato (George et al., 2014) George et al. ( 2015) also compared the carbohydrate yield of some genotypes between transplanting and direct planting by using seed roots cut in half.Cutting seed roots is known to restrict the degree of enlargement of the mother root (Nakazawa, 1973).A seed root planter for direct planting cultivation in sweet potato was newly developed to reduce high labor demands (Matsuo & Ishii, 2021).This machine was designed to plant seed roots without cut in half.Therefore, the results from experiments using seed roots without cutting become more important when we evaluate the root traits of genotypes in direct planting as breeding materials.Summarily, additional multi-year evaluations using whole roots are required to better understand the differences in dry matter productivity of crops grown by transplanting and direct planting.
In this study, we evaluated the effect of sweet potato genotypes on DMC and dry matter yield in both transplanting and direct planting for a period of two years (2020)(2021).Based on these results, we aimed to identify the correlation between root DMC and root dry matter weight (DMW) of genotypes grown by the two planting methods.Through these examinations, it should be possible to effectively utilize the accumulated knowledge of genetic resources grown by conventional transplanting for the breeding in direct planting.

Plant materials and growth conditions
Field experiments were performed in 2020 and 2021 at the Kyushu-Okinawa Agricultural Research Center, NARO, Miyakonojo, Miyazaki Prefecture, Japan (31°45 N, 131°00 E), using 28 genotypes (Table 1).The 28 genotypes were identical to those reported by Sakaigaichi et al. (2022).Five of the genotypes (Nos.1-5 in Table 1) were derived from a direct planting breeding program, known to produce mother roots with less enlargement.The remainder 23 genotypes (Nos.6-28 in Table 1) were cultivars or landraces that were selected without considering direct planting suitability.
In the transplanting plots, stem cuttings with five nodes were prepared at nursery beds and transplanted to fields on 24 April 2020, and 23 April 2021.Direct planting was performed using seed roots that had been harvested in the prior fall and kept in a storehouse.Seed roots without cutting were directly planted in the fields on 18 March 2020, and 24 March 2021.Seed roots weighing 40-177 g in 2020, and 52-167 g in 2021, were used.In both transplanting and direct planting conditions, each plot contained three serial hills at a density of 4.04 hills m −2 (interhill space, 33 cm) with one replication.Chemical fertilizer (4 g N m −2 , 6 g P 2 O 5 m −2 , and 10 g K 2 O m −2 ) was applied as a basal dressing each year.

Harvesting and sample analysis
Harvest surveys were conducted on 31 August 2020, and 25 August 2021, for both the transplanting and direct planting plots.During each harvest, all storage roots were dug from the plots using machines.In the direct planting plots, the storage roots were sorted into mother and daughter roots.All the roots were cut into small pieces using a cutting machine.Sub-samples, with 100 g fresh weight, were prepared in each plot and well dried at 80°C to determine the root DMC (%, fresh matter).Root DMW in the transplanting plot was calculated from the product of the root fresh matter weight and DMC.In direct planting, the fresh matter weight of mother and daughter roots was multiplied by their root DMCs, respectively, and these products were summed to evaluate the total DMW.Here, the fresh matter weight of mother and daughter roots in direct planting plots was referred from the data of our previous study (Sakaigaichi et al., 2022).

Meteorological data
Data on solar radiation and temperature during this experiment at the experimental site were obtained by using the Agro-Meteorological Grid Square Data (Ohno et al., 2016).

Statistical analysis
The correlation between root DMC in transplanting and mother and daughter root DMCs in direct planting was evaluated for each year by calculating the Pearson's correlation coefficient (r).The correlation between the root DMW in transplanting and the total root DMW, which is the sum of the mother and daughter root DMWs in direct planting, was also calculated.All statistical analyses were performed using the SPSS ver.21.0 (IBM, Armonk, NY, USA).

Results and discussion
The solar radiation and daily average temperature during this experiment are shown in Figure 1.The level of solar radiation from March to August 2020 was higher than that of 2021, except in July.The daily average temperatures in 2020 and 2021 fluctuated similarly from March to July; however, the temperature in August 2021 was lower than that in August 2020.According to Tanaka et al. (2018), the starch yield grown at this experimental site was significantly affected by sunlight hours and temperature, especially in August.In this study, the mean root DMW of the 28 genotypes grown by transplanting was higher in 2020 (248 g) than that in 2021 (193 g) (Table 1).As with the root DMW in transplanting, the mean total root DMW of the 28 genotypes grown by direct planting was also higher in 2020 (254 g) than that in 2021 (233 g) (Table 1).These results suggest that the weather conditions in 2020 were preferable to those in 2021 for the yield productivity of the sweet potato.
Among the 28 genotypes, root DMC in transplanting had no correlation with mother root DMC in direct planting in 2020 (r = 0.270, p = 0.165; Figure 2(a)).On the other hand, there was a significant correlation between root DMC in transplanting and mother root DMC in direct planting in 2021 (r = 0.515, p < 0.01; Figure 2(b)).Unlike in the case of mother root DMC, root DMC in transplanting was significant highly correlated with daughter root DMC in direct planting in both 2020 (r = 0.842, p < 0.001; Figure 2(c)) and 2021 (r = 0.948, p < 0.001; Figure 2(d)).Root DMC in transplanting and daughter root DMC in direct planting were plotted around the 1:1 ratio line in almost all genotypes (Figure 2(c, d)).As mentioned above, the yield productivity of sweet potato was higher in 2020 than that in 2021 because of preferable weather conditions (Table 1, Figure 1).Considering these results, the close correlation between root DMC in transplanting and daughter root DMC in direct planting could be retained independently of different yields such as in 2020 and in 2021.
As previously described, only daughter roots can be shipped as products because the mother roots have no Icommercial value.Therefore, we conclude that root DMC in conventional transplanting is a practical and reliable indicator for predicting daughter root DMC in direct planting.DMC of sweet potato is determined mainly by the additive effect of polygenes (Baafi et al., 2017;Sakai, 1964).As such, crossing high DMC genotypes can promote the accumulation of genes related to high DMC in the offspring, generating subsequent generations of cultivars with high DMC.In fact, cultivars with high DMC and starch content, such as 'Shiroyutaka' (Sakamoto et al., 1987), have been released in Japan through this breeding strategy.We believe that broadening the breeding material to high DMC genotypes that were already screened in conventional transplanting should accelerate the DMC of daughter roots of breeding lines in direct planting programs.
ICorrelation coefficients between root DMC in transplanting and mother root DMC in direct planting were smaller in both years when compared with the coefficients between root DMC and daughter root DMC (Figure 2(a-d)).Unlike in the case of daughter root DMC, the correlation between root DMC in transplanting and mother root DMC in direct planting was not significant in 2020 but significant in 2021 (Figure 2(a,b)).Our previous study (Sakaigaichi et al., 2022) suggested that mother-root traits (e.g.mother root weight) had lower heritability than daughter root traits (e.g.daughter root weight).These results imply that the mother root DMC in direct planting is more affected by the cultivation environment than the daughter root DMC.It is difficult to infer the mother root DMC in direct planting from the root DMC in transplanting.
Regarding dry matter yield, root DMW in 2020 ranged from 61 g ('Bis7-1') to 492 g ('Murasakimasari'; Table 1), and in 2021 ranged from 60 g ('Amaya') to 311 g ('Shiroyutaka').Our previous study (Sakaigaichi et al., 2022) highlighted a large variation in the ratio of daughter root to mother root weights among the 28 genotypes grown by direct planting.Therefore, the total root weight, which is the sum of the daughter root and the mother root weights might be considered as the yield potential of each genotype in direct planting programs.The total DMW in direct planting was compared with the root DMW in transplanting.The total root DMW including mother and daughter roots in 2020 ranged from 28 g ('Yen500') to 554 g ('Suzukogane'; Table 1), and in 2021 ranged from 76 g ('Amaya') to 446 g ('Hawaii No.2').Genotypes of Nos.1-5 in Table 1 were derived from direct breeding program; 'Suzukogane'(No.2) and 'Kyushu No.198'(No.3)indicated the relatively high levels of total root DMW in direct planting when compared with the performance of 'Koganesengan' (No.8) and 'Shiroyutaka'(No.9)(Table 1).
Among the 28 genotypes, the root DMW in transplanting had a significantly high correlation with the total root DMW in direct planting in both 2020 (r = 0.730, p < 0.001; Figure 3(a)) and 2021 (r = 0.673, p < 0.001; Figure 3(b)).As shown above, the total root DMW in the direct planting was calculated from the sum of the mother and daughter root DMWs.Therefore, the total root DMW can be considered as the dry matter yielding potential for each genotype.Sweet potato crops grown by direct planting are different from those grown by transplanting, such as stand establishment (George et al., 2011;Sakaigaichi et al., 2021) and partitioning of photosynthetic products between mother and daughter roots (George et al., 2011;Kobayashi, 1968).Additionally, in this study, the weather conditions were different between 2020 and 2021, and the conditions were preferable in 2020 (Table 1, Figure 1).Even under these complicated circumstances, our study identified significant correlations between root DMW and total root DMW for two years (Figure 3(a,b)).
In contrast to DMC and starch content, root yield is controlled by non-additive effects of genes, mainly heterosis effects (Sakai, 1964).Therefore, to increase yield productivity, it is also important to find the crosses that show high combining ability while paying attention to the inbreeding depression of the yield (Yoshida, 1986).
As mentioned above, the breeding program for direct planting is unique because it requires high attention to the partitioning of photosynthetic products between the mother and daughter roots.However, this study revealed that the genotypes' DMC of the daughter root and total root DMW in direct planting could be presumed even from existing breeding records in conventional transplanting cultivation.Unlike the previous report (Kobayashi, 1972), we clarified from the multiyear evaluation that these close correlations could exist stably even when the meteorological conditions that affect yield productivity are different.These findings largely reduce our efforts to screen breeding material for elevating root DMC in direct planting, and those contribute to broadening the genetic basis for the breeding of direct planting in Japan.
matter content; DMW, dry matter weight; Total root, sum of mother and daughter rootsGenotypes of Nos.1-5 and those of Nos.6-28 indicate the genotypes derived from the direct planting breeding program and from other sources, respectively.The values in parentheses indicate the mean for genotypes derived from direct planting (No.1 to 5).

Figure 2 .
Figure 2. Correlation of root dry matter between transplanting and direct planting cultivations.DMC, dry matter content Graph A and B indicate the relationships between Root DMC in transplanting and Mother root DMC in direct planting in 2020 and 2021, respectively.Graph C and D indicate the relationships between Root DMC in transplanting and Daughter root DMC in direct planting in 2020 and 2021, respectively.Black and white symbols (Nos.1-5 and 6-28, respectively) indicate the genotypes derived from the direct planting breeding program and from other sources.r, Pearson's correlation coefficient; **, and *** indicate significant differences at 1% and 0.1% of probability, respectively.Broken lines describe 1:1 ratio.

Figure 3 .
Figure 3. Correlation of root dry matter weight between transplanting and direct planting cultivations.DMW, dry matter weight; Total root, sum of mother and daughter roots Graph A and B indicate the relationships between Root DMW in transplanting and Total root DMW in direct planting in 2020 and 2021, respectively.Black and white symbols (Nos.1-5 and 6-28, respectively) indicate the genotypes derived from the direct planting breeding program and from other sources.r, Pearson's correlation coefficient; *** indicates significant differences at 0.1% of probability.Broken lines describe 1:1 ratio.

Table 1 .
Root dry matter content and root dry matter weight of 28 sweet potato genotypes cultivated by transplanting and direct planting.