Transformation Capability Optimization and Product Application Potential of Proteatia brevitarsis (Coleoptera: Cetoniidae) Larvae on Cotton Stalks

Simple Summary The Xinjiang Uyghur autonomous region is the most important area for cotton production in China, where recycling of cotton stalks (CS) as a useful resource should be encouraged. This article investigated the technical feasibility of CS as a feed and fertilizer based on the transformation of P. brevitarsis larvae. Decomposition inoculant, fermentation duration, and cattle manure ratio were considered the key factors affecting the transformation capability of P. brevitarsis larvae on CS. The research showed that 40–50% of cattle manure, 0.1% VT inoculant, and a fermentation duration of 25–30 days were the optimal technical parameters. The protein content of the larval body was as high as 52.49%, and the fat content was 11.7%. The organic matter content of frass (larvae dung-sand) was 54.8%, and the content of total nitrogen, phosphorus, and potassium (TNPK) was 9.04%, which is twice more than that of the organic fertilizer standard (NY525-2021, Beijing, China, TNPK ≥ 4.0%). The application of CS as feed (larval body) and fertilizer (larvae dung-sand) is feasible, promoting the utilization of both CS and cattle manure. Abstract Cotton stalks (CS) are a potential agricultural biomass resource. We investigated the use of CS as a feed for Proteatia brevitarsis Lewis larvae and the resulting frass (larvae dung-sand) as a fertilizer. Based on a three-factor experiment (decomposition inoculant, fermentation duration, and cattle manure ratio), the optimal parameters for the transformation of CS using P. brevitarsis larvae were determined as 40–50% of cattle manure, the use of VT inoculant and a fermentation duration of 25–30 days. Regarding the products of the transformation, the protein content of the larval body was as high as 52.49%, and the fat content was 11.7%, which is a suitable-quality insect protein source. The organic matter content of larvae dung-sand was 54.8%, and the content of total nitrogen, phosphorus, and potassium (TNPK) was 9.04%, which is twice more than that of the organic fertilizer standard (NY525-2021, Beijing, China, TNPK ≥ 4.0%), and larvae dung-sand has the potential of fertilizer application. Therefore, CS as a feed and fertilizer based on the transformation of P. brevitarsis larvae is feasible, and it is a highly efficient way to promote the utilization of both CS and cattle manure.


Introduction
The Xinjiang Uyghur autonomous region is the most important area for cotton (Gossypium hirsutum L.) production in China. The cotton planting area is about 2.5 million hectares, and the cotton yield exceeds 5.0 million tons [1]. This area also produces cotton stalks (CS) equivalent to five times the cotton yield. Excluding the cotton leaves and root stubble, the this study is to provide a method reference for improving the transformation capability of organic waste and promoting the utilization of cotton stalks and cattle manure.

Experimental Site
The experimental site was located in the Industrialization Research Base of Environmental Insect Transforming Organic Waste, Xinjiang Agricultural University, in Manas County (44 • 13 49 N, 86 • 23 3 E), Changji Prefecture, China.

Experimental Materials
Cotton stalks (CS) and cattle manure were taken from farmers or herders around the base. The larvae of P. brevitarsis were self-reproduced in the base. Materials such as decomposition inoculants (Table 1), cucumber (Cucumis sativus L.) seeds (Changchun Mithorn, Xinjiang Lianchuang Seed Co., Ltd., Urumqi, China; for the determination of the seed germination index), electronic balance (LT3002, Changshu Tianliang Instrument Co., Ltd., Changshu, China) and experimental tools were purchased or previously owned. CS and cattle manure were dried and crushed for use. The three-factor five-level orthogonal experiment (Table 2) of decomposition inoculant, cattle manure ratio and fermentation duration were conducted in September 2020. A total of 25 treatments were designed by IBM SPSS Statistics 23.0 (SPSS 23.0) (L25 (5 6 ) orthogonal table) and recorded as A 1-5 B 1-5 C 1-5 . The CK groups were the CS fermented for 0, 10, 15, 20, 25, and 30 days. The initial materials for every treatment were 90 kg (dry weight, the same as below). The decomposition inoculants were added at the recommended amount. The water content (WC) of the materials was adjusted to 65 (±5)%. Then, the materials were mixed and piled into a cone shape. The ambient temperature and fermentation temperature of material pile (20 cm depth) were recorded daily. Samples were taken from 20 to 30 cm below the surface of material pile (five-point sampling method) according to the days of fermentation duration for each treatment. Each sample weighed 3 kg (fresh weight) and was frozen and stored in the refrigerator. In strict accordance with the process of turning the material pile every 5 days and sampling first and then turning the pile, and the material fermentation and sampling experiments were finished after 30 days. The samples were thawed naturally, and each culture box (1 L) was filled with 280 g of fresh material (about 80 g dry weight), 10 larvae (the 3rd instar and 15th day) of P. brevitarsis were put into the box. Thereafter, the transformation experiment was carried out for 15 days. Each treatment was repeated four times. On the 16th day, weighing larvae weight gain, feed intake and dung-sand weight, the feed utilization rate, dung-sand conversion rate and mortality were calculated by Liu (2012) [80]. The optimum technical parameters were selected by making a comprehensive comparison of the transformation capability of larvae.

Validation of the Optimal Technical Parameters for CS as Feed and Fertilizer
The validation experiment was carried out in May 2021. The optimal combination based on the experimental results of Section 2.3.1 was A 5 B 4 C 4 : VT inoculant, the ratio of cattle manure was 40%, and the fermentation duration was 25 days. The control feed (CK) was cotton stalks fermented for 25 days, and the specific operation is referred to in Section 2.3.1. Thereafter, we determined the transformation capability data of the P. brevitarsis larvae to CS and verified the feasibility of the optimal technical parameters.

Determination of Related Nutritional Indicators for CS Transformation Products as Feed and Fertilizer
The feed or fertilizer nutrition indicators of the raw materials (CS and cattle manure), fermented materials (fermented CS and A 5 B 4 C 4 feed), and products (dry larvae and larvae dung-sand) of the optimal treatment and control were determined (refer to GB 13078-2017 and NY525-2021 standards, Beijing, China, and tested by Sichuan Weil Testing Technology Co., Ltd., Chengdu, China. The seed germination index was determined by referring to the appendix of NY525-2021, Beijing, China). To explore the application potential of CS transformation by P. brevitarsis.

Data Processing
SPSS 23.0 was used to conduct a three-factor five-level analysis of variance with repeated observations and no interaction. One-Way ANOVA was performed for the CK groups and the three factors, and Tukey's multiple comparison analysis was performed for the differences between different treatments (p < 0.05). Microsoft Excel 2013 was used to record and organize data and draw tables. Sigma Plot 14 was used to draw graphs.

Preliminary Selection of the Optimal Combination of Decomposition Inoculant, Fermentation
Duration, and Cattle Manure Ratio 3.1.1. Effect of Fermentation Duration on Transformation Capability to CS Using P. brevitarsis Larvae As shown in Table 3, the transformation capability of the P. brevitarsis larvae on CS was significantly different under different fermentation duration. The optimal indexes of feed intake, larvae weight gain, and feed utilization rate were 25 days after fermentation. The dung-sand weight was the best after 20 days of fermentation, but the difference was insignificant compared with 25 days of fermentation. The dung-sand conversion rate was optimal after 15 days of fermentation, which was not significantly different from that after 20 days of fermentation. The mortality of larvae was the lowest at the 15 and 25 days of fermentation duration, and there was no significant difference among all treatments. Comprehensive analysis showed that the transformation capability of the P. brevitarsis larvae on CS was the best for 25 days after fermentation.

Influence of Three Factors on the Fermentation Temperature of Materials
As shown in Table 4, under the fermentation cycle of every 5 days, the influence of the decomposition inoculant on the fermentation temperature of the material pile did not reach a significant difference level, and the overall situation was relatively stable. The influence of the ratio of cattle manure on the fermentation temperature of the material pile reached a significant difference level on the 10th, 20th, and 30th days. In the first 20 days, the fermentation temperature of the material pile at the 10% cattle manure group was the highest, and that of the 50% cattle manure group was lower. After 25 days, the temperature showed an opposite trend. In terms of fermentation duration, only 25 days of fermentation showed a significant difference level, which should be the inflection point of material fermentation temperature. After 30 days of fermentation, except for the CK group, the fermentation temperature of 25 treatments was above 30 • C, which was much higher than the ambient temperature on the same day. In the early stage, the temperature of the CK group was high, but the temperature dropped sharply after 20 days. The temperature of the 25 treatments only dropped significantly after 25 days of fermentation, which was related to the degree of material fermentation entering the later stage and also related to the low ambient temperature (the average temperature after 20 days was lower than 10 • C). The trend of temperature variation among different treatments showed that adding decomposition inoculant and cattle manure could maintain the temperature of the material pile in a high and stable range and then promote the fermentation of CS.  Table 5 has shown that the transformation capability of the P. brevitarsis larvae with different decomposition inoculants was significantly different in the indexes of feed intake and weight gain but not significantly different in the other four indexes, and VT inoculant was the best. As for the factor of cattle manure ratio, 40% and 50% groups showed the best performance, and the indexes of feed intake, dung-sand weight, feed utilization rate, and dung-sand conversion rate were significantly different from the 10% and 20% groups. The transformation capability of the P. brevitarsis larvae was the best at 25 days and 30 days after fermentation, and the feed intake, dung-sand weight, and feed utilization rate of the third instar larvae were significantly higher than those at 10 days after fermentation. The difference in transformation capability of the larvae under the three factors provided suitable support for optimizing the technical parameters of the transformation of CS using the P. brevitarsis.

Test of Inter-Subjects Effects under Three Factors
It can be seen from Table 6 that the effects of the three factors on feed intake, dung-sand weight, feed utilization rate, and dung-sand conversion rate were significantly different, while the differences in larvae weight gain and mortality were not significant. This experiment mainly analyzed four indexes with significant differences. According to the comparison of the type III sum of squares, the order of influencing factors for the feed intake was from largest to smallest: B > C > A. For the three assessment indicators of dung-sand weight, feed utilization, and dung-sand conversion rate, the order of the three effect factors was C > B > A.

Intuitive Analysis and Tukey Test under Three Factors
As can be seen from Figure 1, when the feed intake (a) and dung-sand weight (b) were used as the screening indicators, the optimal combination of the decomposition inoculant (A), cattle manure ratio (B), and fermentation duration (C) was: VT inoculant, 40% (50%) of cattle manure ratio, and 30 days of fermentation duration. According to the results of intuitive analysis and Tukey's test (Figure 1), and referring to the results that the transformation capability of the P. brevitarsis larvae was the best when CS was fermented for a duration of 25 days (Table 3), the principles of minimizing cattle manure ratio, shortening fermentation duration, and reducing treatment cost were also considered. The optimal combination was A5B4-5C4-5 (0.1% VT inoculant, 40-50% of cattle manure ratio, and 25-30 days of fermentation duration), and A5B4C4 was given preference.

Validation of the Optimal Technical Parameters for the Transformation of CS Using P. brevitarsis Larvae
CS fermentation and transformation experiments were performed under the optimal combination (A5B4C4). The results are shown in Table 7.  When the feed utilization rate (c) was used as the screening indicator, the optimal combination of the decomposition inoculant, cattle manure ratio, and fermentation duration was: VT inoculant, 40% (50%) of cattle manure ratio, and 30 days of fermentation duration. When the dung-sand conversion rate (d) was used as the screening indicator, the RW and NFK inoculant, 40% of cattle manure ratio, and 30 days (25 days) of fermentation duration were optimum.
According to the results of intuitive analysis and Tukey's test (Figure 1), and referring to the results that the transformation capability of the P. brevitarsis larvae was the best when CS was fermented for a duration of 25 days (Table 3), the principles of minimizing cattle manure ratio, shortening fermentation duration, and reducing treatment cost were also considered. The optimal combination was A 5 B 4-5 C 4-5 (0.1% VT inoculant, 40-50% of cattle manure ratio, and 25-30 days of fermentation duration), and A 5 B 4 C 4 was given preference.

Validation of the Optimal Technical Parameters for the Transformation of CS Using P. brevitarsis Larvae
CS fermentation and transformation experiments were performed under the optimal combination (A 5 B 4 C 4 ). The results are shown in Table 7. Table 7. Transformation capability of the 3rd instar larvae of P. brevitarsis under the optimal combination.

Treatments
Feed Intake (g) Larvae Weight Gain (g) It can be seen from Table 7 that under the optimal technology combination, the transformation capability of the P. brevitarsis larvae on the A 5 B 4 C 4 feed was significantly different in feed intake, dung-sand weight, feed utilization rate, dung-sand conversion rate, and mortality with that of CK, and the feed utilization rate and dung-sand conversion rate were over 80%. Therefore, the optimal technical parameters for CS resource utilization were determined as A 5 B 4 C 4 : 0.1% VT inoculant, 40% of cattle manure ratio, and 25 days of fermentation duration. The fresh weight of fermentation material (A 5 B 4 C 4 feed) was weighed, the water content was measured, and the yield of the material was calculated to be 62.85%. It can be concluded that 104.75 g of A 5 B 4 C 4 feed can be obtained by adding 66.67 g of cattle manure for every 100 g of CS raw material. A total of 70.92 g of larvae dung-sand can be obtained by the third instar larvae of P. brevitarsis, and the weight gain of the dry larvae is 3.06 g, and 20.88 g of residue is left. It can be seen from Table 8 that the protein content of fermented CS increased by 41.9%, the crude fiber content decreased slightly, the content of gross energy (GE) was slightly increased, and the contents of crude ash and water-soluble chlorides increased greatly. The crude protein (CP) content of A 5 B 4 C 4 feed reached 13.18%, which was slightly lower than 14.16% of cow manure and was 1.29 and 1.84 times that of the fermented and unfermented CS. Compared to the fermented CS, the A 5 B 4 C 4 feed significantly reduced crude fiber content, increased the crude ash and water-soluble chloride content, and decreased GE. The content of free gossypol (FG) in fermented materials was about 50% lower than that in raw materials. The FG in the A 5 B 4 C 4 feed was not detected in the larvae of P. brevitarsis (detection limit is 20 mg/kg). The protein (52.49%) and fat (11.7%) content of the P. brevitarsis dry larvae were much higher than those of the A 5 B 4 C 4 feed, while the content of crude fiber was only 6.1%, and the content of water-soluble chloride was lower than that of the A 5 B 4 C 4 feed. The GE (19.20 KJ/g) was intermediate between carbohydrate (17.5 KJ/g) and protein (23.64 KJ/g). The insect-microorganism composite systems can improve the nutrition indicators of CS as a feed, and the larval body was 7.31, 19.50, and 1.16 times higher than that of CS in protein, fat, and total energy and more than 50% lower in FG, and the content of crude fiber is only 1/6 of CS. As shown in Table 9, the organic matter (OM) content of the six materials was above 54%, and the CS was the highest (67%). Their total nutrient (TNPK) content was more than 4.0%. The total nutrient (TNPK) and potassium (TK) content of the A 5 B 4 C 4 feed were 9.04% and 4.44%. For the germination index (GI), the unfermented CS (47.09%) and manure (66.87%) had certain toxicity to seed germination, the GI of the remaining four materials was more than 70%, indicating that it was non-toxic to seed germination, and the GI of fermented CS was 102.88, which could promote the seed germination. The pH value of the six materials ranged from 6.6 to 9.5, and it was neutral to alkaline overall. OM decreased, HAs and GI increased first and then decreased, and TNPK, water-soluble chloride, and pH values increased in the insect-microorganism composite process from raw materials to fermentation materials and then to larvae dung-sand. In addition to pH value, two kinds of fermentation materials and two kinds of larvae dung-sand were in line with the latest standards of organic fertilizers in China in terms of OM, NPK, and GI (NY525-2021, NPK ≥ 4%, DOM ≥ 30%, GI ≥ 70%, pH 5.5-8.5).

Discussion
This study showed that for every 100 g of cotton stalks supplemented with 66.67 g of manure, 104.75 g of A 5 B 4 C 4 feed was obtained, and 70.92 g of dung-sand was obtained after transformation by the third instar larvae of P. brevitarsis. The weight gain of the dry larvae was 3.06 g, and 20.88 g of residue remained. The larvae of the P. brevitarsis had a 27.41-fold ability to transform fermented materials (FCR = weight of feed intake/weight gained), which was nearly six times higher than that of the black soldier fly (FCR = 4.5), and had a higher feed utilization rate (80.07% ± 0.65%) and dung-sand conversion rate (84.55% ± 0.53%) [73]. Compared with other dung beetles, P. brevitarsis are more suitable to perform the ecological function of converting organic waste in concentrated agricultural and livestock areas because of their high reproductive ability and their tendency to gather to lay eggs and feed [34,46,65,81,82]. A previous study showed that the ratio of material surface/volume was positively correlated with the fermentation effect, and future work could improve the transformation capability of P. brevitarsis larvae on cotton stalks by reducing the crushing particle size and other measures [75,83]. Previous studies have only focused on the transformation efficiency of the larvae of P. brevitarsis for fermented material; this study also paid specific attention to the productivity from raw materials to fermented materials. According to the calculation results, the productivity of the A 5 B 4 C 4 feed was 62.85%, which was theoretically higher than the rate of traditional organic fertilizer production methods, as judged by the 25 days required for fermentation duration [70][71][72]84,85]. The productivity of fermentation materials can provide data support for the productivity from raw materials to dry larvae and dung-sand.
Some researchers have shown that long-term feeding of excessive amounts of nondetoxified cotton by-products (e.g., cotton leaves, cottonseed meal, and cotton stalks) to vertebrates can lead to the accumulation of free gossypol (FG) in the fed animals, causing poisoning and acute respiratory distress, anorexia, fatigue, and even death [86][87][88]. This has hindered the application of cotton stalks as fodder. In this study, the contents of FG in cotton stalks, cattle manure, fermented cotton stalks, and A 5 B 4 C 4 feed were 96, 114, 47, and 59 mg/kg. The decomposition of inoculant fermentation can significantly reduce the content of FG, which is consistent with the reduction of FG content in feed through fermentation in previous studies [89][90][91]. Interestingly, no FG was detected in the P. brevitarsis larvae after feeding on the A 5 B 4 C 4 feed, indicating that the FG did not accumulate in the larvae, which may be related to the larvae-degrading FG through feeding and metabolism or the short feeding time. The specific reason is the direction of future research. The insect-microorganism composite systems can undoubtedly reduce the content of FG, and the study of its degradation mechanism may provide a reference for reducing the toxicity of FG in livestock feeding on cotton by-products. The protein and fat content of the larval body were 52.49% and 11.7%. It was a suitable-quality, high-protein, insect-derived feed ingredient [92,93], and the nutrient composition of the larvae of P. brevitarsis was consistent with previous studies [46,48,94]. In conclusion, it is feasible to transform cotton stalks to dry larvae feed.
Organic matter (OM) and total nutrients (TNPK) are the most commonly used indicators for evaluating organic fertilizer. This study showed that the OM and TNPK indicators of cotton stalks and manure met the Chinese organic fertilizer standards (NY525-2021, China), but they cannot be applied directly as organic fertilizers [95,96]. Therefore, the evaluation of whether the materials can be used as organic fertilizers should refer to other indicators, such as the germination index (GI), humic acids (HAs), the number of beneficial microorganisms, and so on [44,49,50,69]. Furthermore, the application effect on crops is the core criterion for evaluating the quality of an organic fertilizer [97][98][99]. The larvae dung-sand obtained in this study was much better than the Chinese organic fertilizer standard in terms of OM, TNPK, and other nutrition indicators. However, the high pH value and water-soluble chloride content may be the reason for the low GI of seeds. The quality of larvae dung-sand as organic fertilizer can be improved by adjusting pH and other measures. On the other hand, larvae dung-sand has the characteristics of regular particles and uniform texture, which is easy to process and use and can be processed into prototype flower fertilizer [31]. In cash crops, it can be applied by sowing while fertilizing or using leaching solution drip irrigation, which has the potential to be used as dung-sand-based organic fertilizer [44,54,69].

Conclusions
The optimum technical parameters for transforming cotton stalks using P. brevitarsis larvae were supplementation with 40-50% of cattle manure, the addition of 0.1% VT inoculant, and a fermentation duration of 25-30 days. The dry larvae are a high-protein feed ingredient from an insect-derived, which can be fed and recycled into the ecological breeding industry. The larvae dung-sand is rich in nutrition and has the potential for fertilizer application. This study preliminarily proves the feasibility of cotton stalk feeding and fertilizer dual-use technology based on the transformation of P. brevitarsis larvae. It possesses substantial significance for both theoretical and practical investigations related to boosting the recycling utilization of cotton stalks and cattle manure.