Introduction

In current ecological and environmental science studies, ecological chemometrics is a prominent topic because it can demonstrate the storage status, supply capacity, use efficiency, and limitations of soil nutrients1,2,3. Soil microorganisms regulate the fixation and mineralization of carbon (C), nitrogen (N), and phosphorus (P) in the ecosystem, maintaining the soil in an optimal state conducive to their growth and development. Microbial biomass stoichiometry can reveal the nutrient limitation status of soil microorganisms4. For instance, (1) The ratio of microbial biomass carbon to nitrogen (MBC/MBN) can indicate the species composition of microorganisms in soil5,6. (2) The ratio of microbial biomass carbon to phosphorus (MBC/MBP) indicates the phosphorus uptake and fixation capacity of soil microorganisms7,8. (3) The ratio of microbial biomass nitrogen to phosphorus (MBN/MBP) reflects the magnitude of the plant's demand for soil nitrogen and phosphorus during a certain period4. Soil extracellular enzyme activity (EEA), that is, C, N, and P acquisition enzymes, are formed by the secretion of complex organic matter by soil microorganisms and plant roots in the environment and can be used to obtain nutrients necessary for life activities9,10. For example, in nutrient-poor environments, soil microorganism’s decompose refractory organic matter in the soil by secreting large amounts of N/P-acquiring enzymes to obtain the nutrients needed for growth11. Soil extracellular enzyme stoichiometry (EES), which combines Ecological Stoichiometry Theory (EST) with the Metabolic Theory of Ecology (MTE), can enhance our understanding of the microbial community's metabolic function and the state of nutrients in the soil5,12,13. Soil microorganisms can regulate their extracellular enzyme production and stoichiometric ratios to preferentially metabolize limited resources14. For example, Jiang et al. found that with the increase of soil TP content, acid phosphatase activity and soil microbial phosphorus limitation decreased significantly15. Therefore, for accurate and efficient management of agricultural production, a precise evaluation of the function of ecological stoichiometry is required16.

One of the most crucial management strategies for crop growth in agroecosystems is fertilization, which significantly enhances the resistance and immunity of agricultural products, increases soil nutrients, and promotes agricultural production17. In the modern agricultural development model aimed at high yield, China has become the world's largest producer and consumer of chemical fertilizers18. Although fertilizer use can enhance crop yield and quality and improve soil nutrient status, modern agricultural practices are unduly reliant on the use of fertilizers to increase crop productivity, which, over time, can lead to soil degradation and fertility depletion19,20,21,22,23. Thus, it is more sustainable to apply a combination of fertilizers to ensure environmental sustainability. In agroecosystems, fertilizers can modify soil microbial activity, increase soil nutrient availability, and affect the soil-microbes-enzymes stoichiometric characteristics24,25,26. Long-term fertilizer application enhances crop yields, increases nutrient availability and microbial biomass in the soil, and alters the stoichiometry of nutrients to some extent27,28. Soil fertility, microbial community composition, and soil enzyme activity are the primary topics of current research on the impacts of fertilization on agroecosystems29,30,31. Nevertheless, few studies have examined the relationships between soil-microbes-enzymes stoichiometry and plant quality.

Pitaya (Hylocereus undulatus) is a plant of the genus Hylocereus in the family Cactaceae of the order Opuntiales and is a climbing fleshy shrub native to Central America. The plant was then imported to Southeast Asian nations and China32,33, where it is now widely cultivated in Hainan, Yunnan, Guangxi, Guizhou, and other regions of China. Large areas of carbonate rocks are exposed in the karst region of China, leading to significant people-land conflicts, economic regression, and high poverty rates due to the irrational intervention of human activity34,35. Hence, the local government and people have widely planted the specialist cash crop—pitaya to promote ecological restoration to combat rocky desertification and foster economic development36,37. Numerous field studies have demonstrated the necessity of supplementing fertilizers to enhance soil conditions, support production, and improve quality to meet pitaya growth requirements38,39,40,41. However, less research has been conducted on how fertilizer use affects pitaya quality and the ecological stoichiometry of soil-microbes-extracellular enzyme interactions.

Therefore, the pitaya variety 'Zihonglong' was chosen as the research subject for this study. This study investigated the effects of fertilization on soil microbial and extracellular enzyme contents and stoichiometric characteristics, and explored the limiting conditions of soil nutrients. Moreover, the effects of fertilization on pitaya quality were analysed a view to propose suitable fertilization strategies for increasing the yield and economic benefits of pitaya in the study area. We formulated the following hypotheses: (1) Fertilizer application can significantly enhance pitaya quality; (2) Fertilizer application will significantly affect soil-microbes-extracellular enzyme contents and stoichiometric characteristics in pitaya orchards; and (3) Soil, microorganisms, and extracellular enzymes under fertilizer treatments will positively affect fruit quality.

Results

Response of pitaya quality to fertilizer application

The fruit quality under the different fertilization treatments during the experiment is shown in Fig. 1. Except for the FSI (fruit shape index), all other indices showed significant differences among treatments. The trends of PFW (per-fruit weight), PPY (per-plant yield), ACY (anthocyanin), RTT (solid-acid ratio), and RST (sugar-acid ratio) followed the order: FM (chemical fertilizer + organic fertilizer) > F (single application of chemical fertilizer) > M (single application of organic fertilizer) > CK (control). The trends of the Vc (vitamin C), TSS (total soluble solids), and SS (soluble sugar) contents followed the order: FM > M > F > CK. Compared with those in the CK treatment, the ACY and TA (titratable acid) contents in the F, M, and FM treatments differed, but these differences were not obvious. The use of chemical and organic fertilizers, alone or in combination, significantly increased pitaya PFW, PPY, and the contents of Vc, TSS, and SS. Overall, the chemical fertilizer combined with organic fertilizer (FM) treatment positively affected all the quality indicators of the pitaya.

Figure 1
figure 1

The content of quality components of pitaya under different treatments. PFW per-fruit weight, FSI fruit shape index, PPY per-plant yield, Vc vitamin C, ACY anthocyanin, TSS total soluble solids, SS soluble sugar, TA titratable acid, RTT solid-acid ratio, RST sugar-acid ratio. CK, control; F, single chemical fertilizer application; M, single organic fertilizer application; FM, chemical fertilizer + organic fertilizer.

Full size image

Soil carbon, nitrogen and phosphorus concentrations and their stoichiometry in response to fertilizer application

Under the different fertilization treatments, the soil nutrient concentrations significantly affected (Fig. 2), and the SOC (soil organic carbon), TN (total nitrogen), and TP (total phosphorus) contents significantly decreasing trend. The SOC content in the control group was significantly greater than that in the F, M, and FM treatment groups, and the SOC content in the FM treatment group was 38.6% lower than that in the F treatment group. The TN content decreased by 7.9%, 15.1%, and 28.6% in the F, M, and FM treatments, respectively, compared to that in the control. The TP content of the control was significantly greater than that of the FM treatment and was 1.1, 1.3, and 1.4 times greater than that of the F, M, and FM treatments, respectively (Fig. 2a–c).

Figure 2
figure 2

Soil element contents and stoichiometry of the different treatments. SOC soil organic carbon, TN total nitrogen, TP total phosphorus. C/N, soil SOC:TN ratio; C/P, soil SOC:TP ratio; N/P, soil TN:TP ratio. CK, control; F, single chemical fertilizer application; M, single organic fertilizer application; FM, chemical fertilizer + organic fertilizer.

Full size image

The different fertilizer treatments had more significant effects on the soil C/N and C/P ratios but not on the N/P ratio. The soil C/N (soil SOC:TN ratio), C/P (soil SOC:TP ratio), and N/P (soil TN:TP ratio) ratios varied in the ranges of 6.0-7.9, 16.0-21.1, and 2.6-2.8, respectively. With the use of fertilizers, soil C/N and C/P showed a significant decreasing trend. Compared with treatments F and M, soil C/N and C/P under the FM treatment decreased by 19.2% to 22.0% and 23.7% to 24.2%, respectively; while soil N/P showed an increasing trend but did not reach a significant level (Fig. 2d–f).

Soil biological properties and their stoichiometry in response to fertilizer application

The contents of MBC (soil microbial biomass carbon) and MBN (soil microbial biomass nitrogen) were significantly affected by the different fertilizer treatments (Fig. 3). Generally, fertilizer application tended to reduce the MBC, MBN, and MBP (soil microbial biomass phosphorus) contents in the soil. However, compared with those under the control treatment, the MBC and MBN contents under treatment F increased by 12.2% and 12.1%, respectively. Although the MBP content did not significantly differ among the fertilizer treatments, it still showed a decreasing trend. Compared with those under treatment F, the MBC and MBN contents under treatments M and FM decreased significantly, ranging from 51.6% to 62.4% and from 33.9% to 44.3%, respectively (Fig. 3a–c).

Figure 3
figure 3

Soil microbial biomass and stoichiometry in the different treatments. MBC soil microbial biomass carbon, MBN soil microbial biomass nitrogen, MBP soil microbial biomass phosphorus. MBC/MBN, soil MBC:MBN ratio; MBC/MBP, soil MBC:MBN ratio; MBN/MBP, soil MBN:MBP ratio. CK, control; F, single chemical fertilizer application; M, single organic fertilizer application; FM, chemical fertilizer + organic fertilizer.

Full size image

The different fertilization treatments had more significant effects on the MBC/MBN (soil MBC:MBN ratio) and MBC/MBP (soil MBC:MBP ratio), but not on the MBN/MBP (soil MBN:MBP ratio). The soil MBC/MBN, MBC/MBP and MBN/MBP ratios ranged from 7.7–11.2, 6.3–11.5, and 0.9–1.2, respectively. Except for the control group, among treatments F, M, and FM, MBC/MBN, MBC/MBP, and MBN/MBP values were the highest under treatment F, which was clearly dominant. However, the difference in MBN/MBP among the treatments did not reach a significant level. In conclusion, the trends of the MBC/MBN, MBC/MBP, and MBN/MBP ratios were approximately the same as those of the soil MBC, MBN, and MBP contents (Fig. 3d–f).

Four enzymes related to soil carbon, nitrogen and phosphorus cycling, βG (β-1,4-glucosidase), NAG (β-1,4-n-acetylglucosaminidase), LAP (leucine aminopeptidase) and ACP (acid phosphatase), generally showed a decreasing trend under the different fertilizer treatments, and there was no significant difference (Fig. 4a–d). For the key enzyme of the carbon cycle, βG, the content in the FM treatment increased compared to that in the control, F, and M treatments by 1.04, 1.09, and 1.08 times, respectively.

Figure 4
figure 4

Soil extracellular enzyme activities and their stoichiometry in the different treatments. βG β-1,4-glucosidase, NAG β-1,4-n-acetylglucosaminidase, LAP leucine aminopeptidase, ACP acid phosphatase. Enz C/N, soil Ln βG/Ln (NAG+LAP) ratio; Enz C/P, soil Ln (NAG+LAP)/Ln ACP ratio; Enz N/P, soil Ln βG/Ln ACP ratio. CK, control; F, single chemical fertilizer application; M, single organic fertilizer application; FM, chemical fertilizer + organic fertilizer.

Full size image

Compared with the control, the Enz C/N (Soil Ln βG /Ln (NAG+LAP) ratio), Enz C/P (Soil Ln (NAG+LAP)/ Ln ACP ratio), and Enz N/P (Soil Ln βG /Ln ACP ratio) showed an overall increasing trend among the treatments, although there were little changes (Fig. 4g–i). Enz C/N under treatments F and M increased significantly by 3.3% and 6.7%, respectively, compared with the control. Except for Enz C/N, the differences in Enz C/P and Enz N/P under the fertilizer treatments did not reach a significant level. Vector length was significant difference and increasing trend, indicating that the microorganisms were more affected by C limitation. Although the variation in the vector angle did not reach a significant level, an angle greater than 45° under all treatments indicated that the microorganisms were strongly P-limited (Fig. 4e,f).

Relationships between soil-microbes-extracellular enzyme ecological stoichiometry and pitaya quality

Soil carbon (C) had positive correlations with nitrogen (N) and phosphorus (P). The correlations between soil N/P and soil extracellular enzyme activities (EEA), microorganisms, and fruit quality did not reach significant levels. Except for soil N/P, the correlations between soil enzyme stoichiometry ratios and soil elemental ratios and microbial biomass ratios were negative; however, the correlations between soil elemental ratios and microbial biomass ratios were positive. The soil C/N was significantly or highly significantly negatively correlated with pitaya per-fruit weight (PFW) and the solid-acid ratio (RTT). The soil MBC/MBP ratio was significantly and positively correlated with the MBN/MBP ratio but negatively correlated with most of the quality indicators, except FSI and TA. The titratable acid (TA) was significantly and positively correlated with soil C/N and C/P but negatively correlated with the other quality indicators of pitaya. The correlations between the fruit shape index (FSI) and the soil-microbes-extracellular enzymes were not significant (Fig. 5).

Figure 5
figure 5

Correlations between soil nutrients, microorganisms, extracellular enzymes stoichiometry and pitaya quality. C/N, soil SOC:TN ratio; C/P, soil SOC:TP ratio; N/P, soil TN:TP ratio. MBC/MBN, soil MBC:MBN ratio; MBC/MBP, soil MBC:MBN ratio; MBN/MBP, soil MBN:MBP ratio. Enz C/N, soil Ln βG/Ln (NAG+LAP) ratio; Enz C/P, soil Ln (NAG+LAP)/Ln ACP ratio; Enz N/P, soil Ln βG/Ln ACP ratio. PFW, per-fruit weight; FSI, fruit shape index; PPY, per-plant yield; Vc, vitamin C; ACY, anthocyanin; TSS, total soluble solids; SS, soluble sugar; TA, titratable acid; RTT, solid-acid ratio; RST, sugar-acid ratio.

Full size image

The overall fit of the model is classified into weak, medium, and strong categories according to GoF cut-offs of 0.1, 0.25, and 0.36, respectively42. In this study, the GoF of our model was 0.6, indicating the model fits well. Additionally, there was no mediating effect, and the indirect effect T value was <1.96. The factor loadings of most of the observed variables in the outer model were >0.7, indicating a high degree of convergence for this model. Partial least squares-structural equation model (PLS-SEM) analysis revealed the relationships among soil nutrients, microorganisms, extracellular enzyme activity, and pitaya quality under fertilizer application (Fig. 6). The analyses revealed that microorganisms had a direct positive effect on enzyme activity, soil nutrient content, and pitaya quality. Soil nutrients had a direct positive effect on enzyme activity, while soil nutrient content and enzyme activity had a negative impact on pitaya quality. PLS-SEM demonstrated that the direct and indirect effects of microorganisms and soil nutrients on pitaya quality were inconsistent.

Figure 6
figure 6

PLS-SEM of the relationships between soil-microbes-extracellular enzymes and pitaya quality. Each rectangular box represents an observed variable; each ellipse represents a latent variable. Path coefficients are calculated after 1000 bootstraps and are reflected in the width of the arrow, with blue and red indicating positive and negative effects, respectively. The dashed arrows show that the coefficients did not differ significantly (P > 0.05).

Full size image

Discussion

This study aimed to evaluate the quality of fruit at the full-fruiting stage of pitaya. The results showed that none of the fertilizer treatments had a significant effect on the fruit shape index of pitaya, while there were some differences in the intrinsic quality of pitaya among the different fertilizer treatments. Fruit quality can be significantly improved by applying either a single chemical or organic fertilizer or a combination of chemical and organic fertilizers compared to the control. This is because fertilization effectively combines the soil's macronutrients with the medium and trace elements, reducing the titratable acid content while increasing the per-fruit weight, vitamin C, and soluble solids (Fig. 1), which is consistent with the findings from a previous study on Ziziphus jujubes43. Therefore, fertilizers (particularly chemical fertilizers combined with organic fertilizers) added to the soil can enhance plant nutrient uptake and utilization by promoting the development of a vigorous and robust underground root system44, which can satisfy the requirements of pitaya growth and development for elements such as potassium, phosphate, and nitrogen. Thus, to achieve the goals of increasing yield and enhancing efficiency, attention should also be given to the scientific proportion of each element in actual production.

The SOC, TN, and TP soil indicators used in this study were chosen to represent the variations in soil nutrients in pitaya orchards under short-term fertilization (Fig. 2). Compared with those in the control treatment, the soil SOC, TN, and TP contents in the FM treatment tended to decrease, especially in the FM treatment (chemical fertilizer + organic fertilizer). Possible reasons include the following: (1) the application of exogenous chemical fertilizers accelerated the decomposition of stable organic carbon fractions in the soil, converting the stable carbon pool into the active carbon pool45. This enabled the crop to use the carbon pool more quickly, thereby decreasing the SOC concentration. (2) The production and quality of pitaya were significantly improved by combining chemical and organic fertilizers (Fig. 1), which led to accelerated depletion of soil nutrients without timely supplementation of exogenous nutrients. (3) Karst soils have low levels of phosphorus and nitrogen46, and fruit maturation and harvesting largely deplete the soil TN and TP contents. The present results showed that the TP content was significantly impacted by each fertilization treatment, but the change was not significant. This is primarily because phosphorus (P) is a sedimentary cyclic element with strong stability that is primarily influenced by the parent material that forms the soil47.

The soil C/N ratio in this study ranged from 6.0 to 7.9, which is lower than China's average value of 14.348. The rate at which soil organic matter decomposes is inversely correlated with the soil C/N ratio49. In other words, the FM treatment with the lowest value (6.0) had the highest rate of soil mineralization, implying that it is better for plant roots to absorb nutrients from the soil when chemical fertilizer is combined with organic fertilizer. The variation in the soil C/P ratio in this study ranged from 16.0 to 21.1, which is significantly lower than the average soil C/P ratio in China (136)48. When the soil C/P ratio is less than 200, soil microorganisms have greater potential to release P from mineralized soil organic matter, and soil MBP has a complementary effect on the soil available phosphorus pool50. This reveals that the soil at the experimental site is mainly limited by carbon (C), which is consistent with Fig. 4. Each fertilizer treatment tended to generally decrease the soil C/N and C/P ratios compared to those of the control, indicating that fertilization accelerated the conversion of soil C to N and P, aggravating C limitation51. The range of soil N/P in this study was 2.6 to 2.8, which is lower than China's average value (9.3)48. This is because the main production area of pitaya in Guizhou Province was originally P-limited, the soil is relatively infertile, and the N nutrient content is poor. In addition, nitrogen exists in the surface soil in a dissolved state, making it more susceptible to loss and leaching29,52.

As the most active components of soil organic matter, MBC, MBN, and MBP are very sensitive to changes in the soil environment and can accurately reflect changes in soil C and N contents53,54. This study revealed that, with the exception of the single chemical fertilizer application, microbial biomass under the other fertilization treatments showed a significant declining trend over a short period. This is because chemical fertilizers are fast-acting nutrients that can quickly increase soil MBC and MBN contents over a short period. However, during the current pitaya fertilization and harvesting period, the research location experienced higher temperatures, higher light levels, and less precipitation (too arid) (Table S1). Therefore, the loss of soil moisture directly led to the dehydration of some microorganisms, reducing overall microbial activity and even resulting in the death and decomposition of some microorganisms55. Organic fertilizer is rich in a variety of nutrients, which are released more slowly than chemical fertilizer, making it difficult for organic fertilizer to be fully degraded and released in a short period of time; therefore, organic fertilizer cannot play a timely role in supplementing soil nutrients56. Moreover, a minor amount of weed growth in the study area may have consumed some of the soil nutrients, degrading soil carbon, nitrogen, and phosphorus concentrations, as well as soil microbial biomass57,58.

The soil MBC/MBN, MBC/MBP, and MBN/MBP ratios determine the direction of microbial activity and the release of organic matter nutrients59. In this study, the MBC/MBN, MBC/MBP, and MBN/MBP ratios ranged from 7.7~8.7, 6.3~13.6, and 0.9~1.2, respectively. These values were higher than the mean value of MBC/MBN (7.6) but lower than the mean values of MBC/MBP (70.2) and MBN/MBP (5.6) in Chinese soils60. This indicates that the soil in this study has a low MBN content and low nitrogen (N) element bioavailability. This is because the harvested pitaya fruits removed nutrients from the soil61, thereby reducing the amount of nitrogen (N) in the soil and limiting MBN fixation (Fig. 3). MBC/MBN ratios can characterize the species composition of microorganisms in the soil5. The MBC/MBN ratios of the various fertilizer treatments were greater than 5 in this study, indicating the dominance of the fungal community. In comparison to the control, fertilizer application generally reduced the soil MBC/MBN, MBC/MBP, and MBN/MBP, especially when chemical fertilizer was combined with organic fertilizer. This illustrates that the addition of organic fertilizer introduced organic nitrogen (N) and phosphorus (P) sources to the soil, satisfying the microbial demand for nitrogen and phosphorus62, altering the structure of the soil microbial community and resulting in significant changes in soil microbial stoichiometry. MBC/MBP ratios indicate the uptake and sequestration capacity of soil microorganisms for phosphorus (P) and generally range from 7 to 3063. However, the MBC/MBP ratios were lower under the fertilizer treatments than under the control (Fig. 3), suggesting that pitaya requires a large amount of nutrients during the full-fruiting period to support its production. This prompted the microorganisms to mineralize more soil organic matter for their energy supply8. Chemical fertilizers combined with organic fertilizers further enhanced the microbial release of soil available phosphorus, which is consistent with the results of Chun-Yue Li in 202064.

In this study, four hydrolases (βG, LAP, NAG, and AP), which are closely related to microbial nutrient metabolism, were examined. These enzymes are directly involved in soil carbon, nitrogen, and phosphorus transformation processes, and they catalyze the production of bioavailable terminal monomers65. None of the four hydrolases and their enzyme stoichiometries exhibited significant changes, which was inconsistent with the findings of earlier fertilization studies. This suggests that shorter fertilizer periods are not sufficient to sharply change the nutrient limitation and nutrient acquisition strategies of microorganisms66. This may be due to the differences in the effects of different types of fertilizers on soil enzyme activities67 and is also impacted by external factors such as soil type, climatic conditions, and field management practices. As observed, there was a downwards trend in overall soil enzyme activities (Fig. 4), attributed to the relatively arid conditions (Table S1). A decrease in available soil moisture impairs soil properties, creating an unfavourable environment for microbial life and leading to the inhibition of microbial activity and a decrease in enzyme activities68,69,70.

Soil extracellular enzyme stoichiometry reflects the nutrient requirements and limitations of soil microorganisms71. The mean values of the extracellular enzyme stoichiometry ratios, Enz C/N, Enz C/P, and Enz N/P in this study were 1.2, 0.9, and 0.7, respectively, which are lower than the global-scale Enz C/N mean (1.41) and higher than the Enz C/P and Enz N/P means (0.62, 0.44)11. This implies that the activities of enzymes related to the carbon and nitrogen cycles (βG, NAG, LAP) are greater, while the activity of enzymes related to the phosphorus cycle (ACP) is lower. Because the soil microorganisms are more restricted by carbon (C) and nitrogen (N). Therefore, the process of competing with plants for energy and nutrients increases the activities of the relevant acquisition enzymes, maintaining the relative balance of their stoichiometry72. A soil Enz C/N ratio of 1:1 indicates an equal rate of carbon (C) and nitrogen (N) mineralization. It may be concluded that in this experiment, soil C experienced a greater rate of mineralization than soil N and P. The soil Enz C/N and Enz C/P ratios were generally greater in the fertilizer treatments than in the control, suggesting that the inputs of fertilizers increased the soil organic matter C mineralizing enzyme activities and promoted soil C mineralization. Compared with those in the other fertilizer treatments, the activities of the four enzymes were lower in the M treatment, and the soil C, N, and P contents were also lower (Fig. 4). This indicates that lower nutrient content leads to weaker extracellular enzyme activities, which is in line with Zuo's conclusion73. Principally, the short study duration, the slow rate of organic fertilizer application, and the fact that the organic matter nutrients within it are not yet functional. Additionally, the enzyme stoichiometry ratios did not reach significant levels in this study, which is inconsistent with the results of other long-term fertilization treatments. This shows that the short duration of fertilizer application is insufficient to sharply alter the nutrient limitation and acquisition strategies of microorganisms74.

The vector angles were all greater than 45°, indicating phosphorus (P) limitation in the study area, with no significant differences among the treatments. The reasons may include the following: (1) The special geological conditions of karst areas in southwest China determine that the soil nutrients are poor, and the phosphorus (P) content is low61. (2) Under short-term fertilization conditions, the effect of each fertilization treatment was not obvious. (3) The soil nutrient output was high during the full-fruiting stage (Fig. 2), and weeds competed for nutrients58. The vector length ranged from 1.4~1.5, indicating that there was a carbon (C) limitation in the study area. This may be caused by the following reasons: (1) The soil in the experimental site is barren and the carbon (C) content is low61. (2) The demand for carbon (C) by soil microorganisms increases during the full fruiting stage, leading to higher values of Enz C/N and Enz C/P in soil and an increased requirement for effective carbon sources by soil microorganisms75. Although the use of fertilizer increased the aboveground carbon (C) input, it was not sufficient to compensate for the reduction in underground carbon input and the decrease in effective carbon sources caused by microbial decomposition76,77. (3) Generally, in P-limited ecosystems, the microbial community tends to consume more carbon (C) to produce and secrete enzymes related to phosphorus metabolism. The microbial demand for phosphorus promotes the metabolism of carbon, resulting in a C-limited on microorganisms in the later period78.

There was a significant negative correlation between the Enz C/P ratio and the soil C/N ratio, C/P ratio, MBC/MBN ratio and MBC/MBP ratio. This is because microbial growth is more limited by phosphorus (P) under conditions of high soil C/P and low nitrogen (N) and phosphorus (P) contents in Karst areas57. Microbial communities adapt by lowering Enz C/P, and our study's results are somewhat consistent with the resource allocation theory79. MBC/MBP was significantly positively correlated with soil C/N and C/P, while MBN/MBP was significantly positively correlated with soil C/P (Fig. 5). These results suggest that soil microorganisms positively impact soil element dynamics (Fig. 6). Although soil microorganisms have a self-balancing mechanism, they respond to soil changes under external stresses to preserve the relative stability of the C:N:P ratio80. Soil C/N and C/P were significantly and positively correlated with pitaya TA but significantly or highly significantly negatively correlated with pitaya PFW, Vc, SS, RTT and RST (Fig. 5). These findings were in line with those of Fig. 6, which showed that soil elements negatively affected the quality of pitaya (path coefficient of -0.823 in Fig. 6). The soil Enz C:N:P ratio was significantly negatively correlated with pitaya TA but positively correlated with the other quality indices (Fig. 5), revealing that EES had a weak negative effect on pitaya quality (path coefficient of -0.077 in Fig. 6). MBC: MBN: MBP was negatively correlated with the quality indices of pitaya (Fig. 5), but the results in Fig. 6 indicate that the effect of microorganisms on the quality of pitaya was weakly positive (path coefficient of 0.218). This is probably because the microbial biomass (MBC, MBN, and MBP) contributed more to the model than did the microbial stoichiometric ratio.

The results of this study were obtained over a short experimental period, and it is difficult to comprehensively summarize the effect of fertilization on the relationships among soil-microbes-extracellular enzymes, and pitaya quality, as fertilization is affected by various conditions, such as the scale of the study and duration. As a result, more specific mechanisms need to be further explored in future research.

Conclusion

In this study, different fertilization treatments had significant effects on the ecological stoichiometric characteristics of soil-microbes-extracellular enzymes, as well as on the fruit quality of pitaya orchards in Karst regions. There were also significant carbon (C) and phosphorus (P) limitations in the experimental area. Pitaya quality was substantially enhanced by fertilization, especially under combined chemical and organic fertilizer treatments, with significant improvements in per-fruit weight, vitamin C, and soluble sugar content. Soil nutrient translocation to plants is high, and large amounts of nutrients are removed through stem pruning and fruit picking, which depletes soil nutrient concentrations and alters ecological stoichiometric characteristics. Therefore, the soil microbe extracellular enzyme contents under the different fertilizer treatments tended to decrease. After applying fertilizer stress, the soil microbial community structure was altered, positively affecting the soil nutrient content, extracellular enzyme activities (EEA), and pitaya quality. Soil elements positively influenced EEA, while soil extracellular enzyme activity and soil nutrients negatively impacted pitaya quality. In conclusion, the results of this study not only revealed the ecological stoichiometric characteristics and nutrient cycling processes of soil in a Karst pitaya orchard but also clarified the correlation between these characteristics and fruit quality. This study provides an important scientific basis for the artificial management and increased fruit yield of pitaya orchards. It is also useful for managing rocky desertification in the Karst areas of Southwest China.

Materials and methods

Experimental site

This research was conducted in Moyang Town, Luodian County, Guizhou Province, China (106°8′42″ E, 25°51′9″ N) (Fig. 7). This region is characterized by a subtropical monsoon climate, with annual precipitation ranging from 1070 to 1200 mm, a mean annual temperature of 23°C, and an annual sunshine duration exceeding 1350 hours. Soils are generally classified as typical Dystric Regosols (China Soil Database, http://vdb3.soil.csdb.cn/). The detailed soil characteristics are presented in Table 1. The plot has been planted with "Zihonglong" pitaya since 2018. From 2018 to 2022, rotten pitaya peel, rhizome, and pomace were used as organic fertilizers for base fertilizer application once a year, according to the local conventional fertilization method.

Figure 7
figure 7

Distribution of the research area in Guizhou.

Full size image
Table 1 The properties of the surface soil (0–20 cm) before the fertilization experiment.
Full size table

Chemical fertilizer and organic fertilizer

In this study, chemical fertilizer and organic fertilizer were selected as the research materials to provide soil nutrients. The chemical fertilizer: branded as Jin Sairui (N:P2O5:K2O content ratio 15:4:26), which was produced by Sichuan Golden-Elephant Sincerity Chemical Co., Ltd. (Sichuan Province, China). (2) Organic fertilizer: branded as, which included Heneng No. 1, total nutrients (N + P2O5 + K2O) ≥ 4%), organic matter ≥ 30%, moisture ≤ 30%, and a pH value of 5.5-8.5, was produced by CGN Leye Bio-Resources Utilization Co., Ltd. (Guangxi Province, China). The organic fertilizer used in this study was a commercial organic fertilizer; the main ingredient was sugar mill filter sludge, with 100% humification, which was made through an aerobic fermentation process and subjected to ex-factory inspection.

Experimental design

Cement columns were used as pillars for the growth of pitaya, with a pile spacing of 3 m × 3 m. There were 10 pitaya seedlings between each pair of concrete pillars, with a plant spacing of 30 cm. The plants were strong, exhibited uniform growth and were of relatively high quality. The experiment comprised four treatments with three completely randomized replication plots: control (CK, no fertilizer), single application of chemical fertilizer (F), single application of organic fertilizer (M), and chemical fertilizer + organic fertilizer (FM), with a ratio of F to M of 1:8. Since this experimental site was located in a karst region limited by geographical constraints and habitat fragmentation, the area of each experimental plot was set to 6 m × 9 m = 54 m2, and the plots were arranged in a completely randomized design, with three replications for each treatment, resulting in a total of 16 plots. The fertilization test period spanned from April to October 2023. Except for the type and amount of fertilization, the other field management practices, such as irrigation, weeding, and pruning, were consistent across all plots. Based on a literature review81 and local routine production surveys, the fertilization content of the applied fertilizers is summarized in Table 2. The fertilizer was applied to a 0-30 cm ditch at the drip line of the pitaya plant.

Table 2 Fertilizer application amount under the different fertilizer application management practices.
Full size table

Fruit sampling analysis

At the end of August 2023, during the fruit maturity period, sample plants with similar tree bodies, consistent growth, and no pests or diseases were randomly selected from each treatment. Nine healthy fruits were collected from each plot, after which the pitaya was peeled, cut into small pieces, and crushed to determine the content of each index.

The vitamin C (Vc) content was determined using high-performance liquid chromatography82. The anthocyanin content (ACY) was analysed using the pH differential method83. The soluble sugar content (SS) was determined using the 3,5-dinitrosalicylic acid method84. The titratable acid (TA) content was determined according to methods proposed by the National Standard of the People’s Republic of China85 using acid-base titration. Total soluble solids (TSS) were measured using a hand refractometer86. Per-fruit weight (PFW) was determined using an electronic balance, and the vertical and transverse diameters of the fruit were measured with a vernier calliper (accuracy: 0.02 mm) and recorded. The per-plant yield (PPY) was the fruit weight of each pitaya plant. The fruit shape index (FSI) was expressed as the ratio of the vertical diameter to the horizontal diameter, the sugar-acid ratio (RTT) was expressed as the ratio of the soluble sugar content to the titratable acid content, and the solid-acid ratio (RST) was expressed as the ratio of the total soluble solids content to the titratable acid content.

Soil nutrient analysis

Soil was sampled after the fruit were harvested, specifically around the end of August and early September 2023. Soil samples were collected from the 0-20 cm layer in each plot, with five points selected from each plot according to the "Sigmoid" curve and combined into one sample. During sampling, we avoided the fertilization area to reduce the effects of fertilization, weeding, and other human interference. The fresh samples were divided into two parts after removing the root systems, the gravel and the animal and plant remains. One part was screened through 2 mm sieves and stored at 4 °C for timely determination of microbial biomass and extracellular enzyme-related indices; the other part was dried naturally, ground, and screened through a 0.25 mm sieve for soil chemical analyses.

The soil pH was determined in a soil: water mixture at 1:2.5 (w/v) using a pH meter (Mettler Toledo, Zurich, Switzerland). Soil organic carbon (SOC) was determined using the K2CrO7-H2SO4 heat method. Soil total nitrogen (TN) was measured using the Kjeldahl procedure. Soil total phosphorus (TP) was detected using the molybdenum antimony resistance colorimetric method. The detailed determination steps for the above soil sample elements are outlined in 'Soil Agrochemical Analysis'87 and were then used to calculate the soil C:N:P stoichiometry. The chloroform fumigation-incubation method88 was used to measure soil microbial biomass C (MBC), N (MBN), and P (MBP), which were then used to calculate the soil microbial biomass C:N:P stoichiometry. Four enzymes important in soil carbon, nitrogen, and phosphorus cycling were selected to determine their activities: β-1,4-glucosidase (βG), acid phosphatase (ACP), β-1,4-N-acetylglucosaminidase (NAG), and leucine aminopeptidase (LAP), which were measured using the microplate fluorescence method89.

Calculations and statistical analysis

The ratio of C-, N-, and P-acquiring enzymes was calculated using Eqs. (1), (2), (3)11:

$$Soil\, enz\, C:N\, ratio = Ln\beta G/Ln (LAP + NAG)$$
(1)
$$Soil\, enz\, C:P \,ratio = Ln\beta G/LnACP$$
(2)
$$Soil\, enz\, N:P \,ratio = Ln (LAP + NAG)/LnACP$$
(3)

Vector analysis (vector length [L, unit-less] and vector angle [A, °]) was used to evaluate microbial nutrient limitation. The vector length and angle were log-transformed and calculated using Eqs. (4), (5)90:

$$Vector\, length\, \left(L\right)=\sqrt{{\left[\frac{Ln\upbeta G}{Ln(LAP+NAG)}\right]}^{2}+{\left(\frac{Ln\upbeta G}{LnAP}\right)}^{2}}$$
(4)
$$Vector\, angle\, (A) = Degrees\, \{ ATAN2\left[ \left(\frac{Ln\upbeta G}{LnAP}\right),\left(\frac{Ln\upbeta G}{Ln\left(LAP+ NAG\right)}\right)\right\}$$
(5)

Soil microorganisms were considered limited by carbon (C) at longer vector lengths. Specifically, when the vector angle was above 45°, soil microorganisms were limited by phosphorus (P); when the vector angle was below 45°, microorganisms were limited by nitrogen (N).

Data calculation and sorting were conducted using Microsoft Excel 2021. One-way analysis of variance (ANOVA) for the ecological stoichiometry of soil-microbes-extracellular enzymes and pitaya qualities was performed using the Tukey test method at a significance level of 0.05 (SPSS 26) and visualized using Origin (2022). Spearman correlation was used to verify relationships among soil nutrients, microbes, and extracellular enzyme C:N:P ratios (Origin 2022). Soil nutrient-related indices were classified as abiotic factors, and the relationships between latent variables were tested using PLS-SEM based on SmartPLS 3. A global criterion for goodness-of-fit (GoF) (computed as \(GoF=\sqrt{\overline{AVE }*\overline{{R }^{2}}}\)) was used to evaluate the model’s overall quality91.