Djulis (Chenopodium formosanum) Stems as Sustainable Sawdust Alternative for Pleurotus sajor-caju Cultivation: A Feasibility Study

Abstract

The heavy reliance of the mushroom industry on sawdust substrates is putting increasing pressure on already limited forest resources, forcing researchers to seek alternative materials. This study investigated the feasibility of using post-harvest djulis (Chenopodium formosanum Koidz.) stems, waste from this indigenous crop in Taiwan, to partially replace sawdust for Pleurotus sajor-caju cultivation. Initial screening with 0–100% djulis replacement revealed growth inhibition above 50% incorporation levels. Refined experiments focusing on 0–30% djulis ratios demonstrated that strain PT exhibited superior adaptation to djulis-containing substrates. Commercial scale grow bag trials showed that among djulis treatments, 25% djulis incorporation achieved the fastest mycelial colonization rate (1.0 cm/day), while 15% incorporation yielded the highest biological efficiency (76.17%), comparable to commercial controls (76.80%). Three-flush harvest cycles confirmed stable productivity across treatments, with total yields ranging from 286 to 320 g/bag. Nutritional analysis showed no major changes in amino acids and antioxidants, with djulis incorporation maintaining protein quality while some enhancement in total free amino acid content and reducing power at 25% incorporation. These findings demonstrate that 15–25% djulis stem substitution sustained commercial production parameters while contributing to sustainable agricultural waste management and reducing forest resource dependence.

1. Introduction

Mushrooms are primarily cultivated using sawdust-based grow bags in Taiwan, with an annual production of approximately 500–600 million bags and a production value exceeding USD 400 million, making a significant contribution to Taiwan’s agricultural sector [1]. However, this success comes with considerable environmental costs, requiring sawdust equivalent to deforesting approximately 2300 hectares of forest and reducing carbon dioxide absorption by about 88,000 metric tons annually [2]. The heavy reliance on forest-derived sawdust has created sustainability challenges that require urgent attention through alternative substrate development.

Among edible mushrooms, Pleurotus sajor-caju (oyster mushroom) is an economically important species primarily cultivated in Taichung and Nantou regions. P. sajor-caju is particularly attractive for cultivation due to its robust growth characteristics, high nutritional value, and ability to grow on diverse lignocellulosic substrates [3,4], making it an ideal candidate for sustainable cultivation systems utilizing alternative substrates.

The mushroom cultivation industry worldwide has increasingly explored agricultural waste materials as alternatives to traditional sawdust substrates. Various studies have demonstrated successful cultivation of oyster mushrooms on rice straw, wheat straw, cotton waste, olive mill effluent, sugarcane bagasse and other agricultural residues [4,5,6,7,8,9]. These alternative substrates not only reduce dependence on forest resources but also provide solutions for agricultural waste management, contributing to circular economy principles.

Djulis (Chenopodium formosanum Koidz.) represents a promising source of alternative substrate material in Taiwan. This traditional crop for indigenous communities has garnered increasing attention for its rich nutritional content and functional components [10,11,12,13]. With a cultivation area exceeding 200 hectares, substantial amounts of plant residues including stems are generated after grain harvesting, currently regarded as agricultural waste and typically plowed back into soil for natural decomposition.

While direct compositional analyses of djulis stems remain limited, studies on closely related quinoa (Chenopodium quinoa) stems provide valuable insights into their potential suitability as fungal substrates. It was reported that quinoa stems contain substantial lignocellulosic components, with carbohydrates comprising 40–50% and lignin approximately 18% of dry matter [14]. Given the taxonomic proximity within the Chenopodium genus, djulis stems likely possess similar lignocellulosic profiles that could support mushroom cultivation. However, djulis stems also contain characteristic saponins and other secondary metabolites, which may influence mushroom growth patterns, necessitating careful evaluation of pre-treatment methods and optimal incorporation ratios.

Substrate composition significantly influences the nutritional and bioactive compound profiles of cultivated mushrooms. Previous research has shown that alternative substrates can enhance certain nutritional parameters, including amino acid content, phenolic compounds, minerals, and antioxidant activities in the fruiting bodies of various mushrooms [15,16,17]. The antioxidant properties of mushrooms, including free radical scavenging activity and reducing power, are particularly important for their functional food applications. These properties are influenced by phenolic compounds, flavonoids, and other bioactive metabolites whose concentrations can vary depending on substrate composition and cultivation conditions. Understanding these relationships is crucial for developing substrates that not only support adequate growth and yield but also maintain or enhance the nutritional value of the final product.

Despite the potential of agricultural residues as mushroom substrates, limited research has been conducted on using djulis stems for P. sajor-caju cultivation. This study aims to explore the use of djulis cultivation by-products as a partial substitute for sawdust in P. sajor-caju cultivation substrates. The research aims to assess the effects of various djulis stem incorporation ratios on mycelial growth, fruiting body yield, nutritional composition, bioactive compounds, and antioxidant capacity. The findings are expected to enhance sustainable reuse of agricultural residues, reduce reliance on forest resources, and address mounting challenges of sawdust shortages and escalating costs in Taiwan’s mushroom industry while potentially improving the functional properties of cultivated mushrooms.

2. Materials and Methods

2.1. Substrate Preparation and Djulis Stem Processing

Fresh djulis stems were harvested, mechanically cut into uniform pieces (0.5–1.0 cm), and thoroughly sun-dried until moisture content reached approximately 10–12%. Prior to substrate formulation, the stems were soaked in distilled water for 8–12 h to facilitate the removal of water-soluble saponins and enhance water absorption capacity. Cultivation substrates were prepared by thoroughly mixing dried sawdust and processed djulis stems with 15% additives (by dry weight) comprising 1% calcium carbonate, 0.5% cane sugar, and 13.5% wheat bran. The moisture content was adjusted to 60–65% using distilled water, and the mixture was homogenized using a mechanical mixer for 10–15 min to ensure consistent distribution.

2.2. Fungal Strains and Inoculant Preparations

P. sajor-caju strains TC and PT were isolated by us from commercial production sites with codes of NPUST-22TC01 (TC strain) and NPUST-22PT07 (PT strain). Fungal isolates were maintained on potato dextrose agar (PDA; Difco, Detroit, MI, USA) at 28 °C. After two weeks of incubation, 1 cm diameter hyphal plugs were excised from the growing margin and used as inoculant for subsequent experiments.

2.3. Small-Scale Mycelial Growth Assessment in Petri Dishes

2.3.1. Initial Screening: 0–100% Djulis Stem Replacement

The prepared substrate mixtures, containing varying proportions of djulis stems as partial or complete replacement for sawdust, were loaded into sterile 9 cm diameter glass Petri dishes. The substrate was carefully compressed to a uniform height of approximately 1.5 cm to ensure consistent density across all treatments. Following high-temperature and high-pressure sterilization (121 °C for 20 min) by an autoclave (YUX-Q40, Yuxin Medical Technology, Taichung, Taiwan), the substrates were allowed to cool to room temperature in a sterile environment. A standardized mycelial plug, obtained from the periphery of a two-week-old colony, was precisely inoculated at the center of each dish. The plates were incubated at a constant temperature of 28 °C in darkness, and radial colony diameter was measured every two days along two perpendicular axes, with the average recorded. Each treatment was replicated five times in a completely randomized design.

2.3.2. Refined Assessment: 0–30% Djulis Stem Replacement

Based on preliminary results indicating optimal performance below 50% djulis stem incorporation, a more detailed investigation was conducted testing the effects of djulis stem incorporation at proportions of 10%, 15%, 20%, 25%, and 30% (designated as DR10, DR15, DR20, DR25, and DR30, respectively) compared to the sawdust control (0% djulis, designated as CK). Experimental procedures followed the same protocol as described above.

2.4. Commercial-Scale Evaluation in Mushroom Grow Bags

2.4.1. Substrate Preparation and Inoculation

Standard commercial-sized mushroom grow bags containing approximately 1.2 kg substrate (65% water content) per bag were prepared using different ratios of djulis stems and sawdust substrates. Following sterilization (121 °C for 60 min), bags were cooled overnight under sterile conditions. The grow bags were manually inoculated at the geometric center of the substrate’s surface with commercially available sawdust spawn of the PT strain of P. sajor-caju at a rate of approximately 50 g spawn per bag. After inoculation, bags were transferred to a dedicated cultivation room maintained at a constant temperature of 28 °C for 35 days during the mycelial colonization phase. During this mycelial growth phase, the bags were kept in complete darkness to promote optimal mycelial development. For each treatment, 24 bags were randomly selected, and the distance between the substrate surface and the leading edge of the mycelial growth was measured every 7 days using sterile measuring tools to compare mycelial growth rates over the 35-day colonization period.

2.4.2. Control Treatments

In addition to the laboratory-prepared sawdust substrate control (CK), a commercially available and certified grow bag (M) was included as a reference control. The commercial grow bag weighed approximately 0.8 ± 0.2 kg and consisted of a proprietary blend of sawdust, rice bran, and calcium carbonate prepared according to industry standards.

2.4.3. Fruiting Body Production and Harvest

Once the mycelium had completely colonized the bags, fruiting induction treatments were initiated. The bags were first subjected to a cold shock treatment at 4 ± 3 °C for 2 days to stimulate primordia formation, after which the cotton plug in the neck of each grow bag was carefully removed under sterile conditions. The bags were then transferred to a climate-controlled fruiting room with precisely maintained daytime temperatures of 28 ± 3 °C, nighttime temperatures of 18 ± 3 °C, and relative humidity maintained at 85–90% using automated humidification systems.

Mature fruiting bodies were harvested when the cap diameter reached 4–5 cm to ensure optimal texture and nutritional quality. For yield assessment, one frame (12 bags) was used to record a single replicate, and each treatment had 12 replicates arranged in a completely randomized design. To ensure harvest quality, misting was discontinued one day before harvesting to prevent excess moisture accumulation on the mushrooms.

Following initial harvest, the surface mycelium on the substrate was monitored for regeneration 10–14 days later. Once the mycelium had fully re-established surface coverage, the bags were re-stimulated to initiate subsequent fruiting cycles. Three complete cropping cycles were conducted, and the fresh weight yield, cumulative total yield, and biological efficiency (%) for each cycle were systematically analyzed. The biological efficiency was calculated using the standard equation: Biological Efficiency (%) = (Fresh weight of harvested mushrooms/Dry weight of substrate at inoculation) × 100.

2.5. Nutritional and Bioactive Composition Analysis

2.5.1. Sample Preparation

Freshly harvested fruiting bodies from each treatment were immediately vacuum freeze-dried (−40 °C for 48 h) by a freeze dryer (Model FD4.5, KINGMECH, New Taipei, Taiwan). Subsequently the samples were ground into fine powder using a high-speed stainless steel grinder mill (Model DM6, YU CHI LING, Changhua, Taiwan) and sieved through a 40-mesh screen to ensure particle size uniformity. All powdered samples were stored in sealed containers at −20 °C until analysis to preserve bioactive compounds.

2.5.2. Basic Nutritional Composition

Analysis of general nutritional composition, including moisture, crude fat, crude ash, crude protein, and total carbohydrates was contracted to the Water Quality Testing Center at National Pingtung University of Science and Technology, Taiwan, following standard methods.

2.5.3. Amino Acid Analysis

Laboratory-conducted analyses included total free amino acids (TFAA), total branched-chain amino acids (BCAA), and lysine content. All amino acid analyses were performed with five replicates per treatment. For determination of BCAA, 100 mg of finely ground mushroom powder was thoroughly mixed with 500 µL of Assay Buffer provided in the commercial kit (Branched chain amino acid colorimetric assay kit (ab83374), BioVision, Milpitas, CA, USA). The mixture was centrifuged at 15,000× g for 10 min at 4 °C by a microcentrifuge (Fresco 17, Thermo Scientific, Dreieich, Germany; hereafter referred to as microcentrifuge) to remove cell debris and other insoluble materials. The clarified supernatant was transferred to sample wells of a 96-well plate, and the volume was adjusted to 50 µL per well with Assay Buffer. Fifty microliters of the prepared reaction mix was added to each well containing the leucine standard and test samples. The contents were gently mixed by pipetting and incubated for 30 min at room temperature, protected from light with aluminum foil. The optical density was measured at 450 nm using a UV-visible spectrophotometer (Spectra Max^® ABSplus, Molecular Devices, San Jose, CA, USA; hereafter referred to as UV-visible spectrophotometer). The BCAA concentration was determined by interpolation from the standard curve prepared with leucine solutions.

For lysine analysis, 0.1 g of freeze-dried tissue sample was homogenized with 1 mL of Extraction Buffer using a vortex mixer. The homogenate was subjected to ultrasonic disruption at room temperature for 5 min (power 20% or 200 W, with 3 s ultrasonic pulses and 7 s intervals, repeated 30 times). The resulting mixture was transferred to a 1.5 mL microcentrifuge tube, sealed, and incubated at 80 °C for 20 min in a water bath. After cooling to room temperature, the sample was centrifuged at 10,000× g for 10 min by a microcentrifuge and the supernatant was carefully collected for the colorimetric reaction. Ten microliters of supernatant was mixed with the working reagent and incubated in a boiling water bath for 5 min. After cooling with running water for 10 s, 170 μL of 60% ethanol was added to the reaction mixture and thoroughly mixed, followed by centrifugation at 5000× g for 10 min at room temperature. The absorbance at 570 nm was measured using a UV-visible spectrophotometer. The lysine concentration was determined by comparison to the standard curve constructed with lysine solutions (CheKine micro lysine assay kit (KTB1470), Abbkine, Wuhan, China).

2.5.4. Phenolic Compound Analysis

The analytical method for determining total phenolic compounds was adapted Sato, Ramarathnam [18] with minor modifications. A 15 mL centrifuge tube was used to mix 0.5 g of freeze-dried mushroom powder with 5 mL of 10 mM sodium phosphate buffer. The mixture was sonicated for 10 min in an ultrasonic bath and then centrifuged at 4 °C at 4000× g for 15 min by a multi-purpose centrifuge (Megafuge 16, Thermo Scientific, Germany; hereafter referred to as multi-purpose centrifuge). An aliquot of 0.1 mL of the clear supernatant was mixed with 2 mL of 20% sodium carbonate (Na₂CO₃) solution, with a reaction time of 2 min. Subsequently, 1 mL of diluted Folin–Ciocalteu’s reagent (10 g H₃PMo₁₂O₄₀, H₃PW₁₂O₄₀ 25 g per liter, pH 1.5) was added, thoroughly mixed, and the reaction was allowed to stand for 30 min at room temperature in darkness. The absorbance was measured at 750 nm using a UV-visible spectrophotometer. A standard curve was prepared using gallic acid at concentrations of 800, 400, 200, 100, and 50 ppm to quantify the total phenolic content of the samples. The experiment consisted of five treatments, each with five replicates.

Total flavonoid content was determined based on the modified method of Van Hung and Morita [19]. Briefly, 0.5 mL of the sample extract was mixed with 1.5 mL of 95% ethanol, 0.1 mL of 10% AlCl₃, 0.1 mL of 1 M potassium acetate, and 2.8 mL of deionized water. The mixture was thoroughly mixed and allowed to react at room temperature for 30 min in darkness. The absorbance was measured at 415 nm using a UV-visible spectrophotometer. A standard curve was prepared using quercetin solutions at concentrations of 20, 40, 60, 80, and 100 ppm to quantify the total flavonoid content of the samples.

2.5.5. Free Radical Scavenging Activity and Reducing Power Assessment

Free radical scavenging activity was determined using the 2,2-diphenyl-1-picrylhydrazyl (DPPH) assay based on the modified method of Shimada, Fujikawa [20]. In this assay, 1 mL of the extract sample was mixed with 1 mL of 0.2 mM DPPH solution prepared in methanol. The mixture was thoroughly mixed by vortexing and allowed to stand for 30 min at room temperature in darkness. The absorbance was then measured at 517 nm using a UV-visible spectrophotometer. For the control, the sample extract was replaced with pure methanol, which contains no antioxidant, and processed under identical conditions. Standard solutions of ascorbic acid (ASA) and butylated hydroxytoluene (BHT) were prepared at concentrations comparable to the test samples and analyzed simultaneously for comparison. The percentage of DPPH radical scavenging activity was calculated using the formula: DPPH scavenging activity (%) = [(Abs control − Abs sample)/Abs control] × 100, where Abs control represents the absorbance of the control and Abs sample represents the absorbance of the sample or standard.

The reducing power assay was conducted based on the modified method of Oyaizu [21]. A 15 mL centrifuge tube was used to mix 0.5 mL of the sample extract with 0.5 mL of phosphate buffer (0.2 M, pH 6.6) and 0.5 mL of 1% potassium ferricyanide. The mixture was thoroughly mixed and incubated in a water bath at 50 °C for 20 min. After cooling for 2 min, 0.5 mL of 10% trichloroacetic acid (TCA) was added, and the solution was centrifuged at 4 °C and 1000× g for 10 min by a multi-purpose centrifuge. Subsequently, 1 mL of the supernatant was transferred to a fresh 10 mL centrifuge tube, mixed with 1.4 mL of deionized water and 0.1 mL of 0.1% ferric chloride solution. The mixture was allowed to stand in the dark for 10 min, and the absorbance was measured at 700 nm using a UV-visible spectrophotometer. Identical concentrations of reference standards, ascorbic acid (ASA) and butylated hydroxytoluene (BHT), were analyzed simultaneously for comparison.

2.6. Data Analysis

All statistical analyses were performed using SPSS Statistics 27.0 software (SPSS Inc., Chicago, IL, USA). For comparisons involving the interaction between strain and substrate composition, a two-way ANOVA was performed. Where significant interactions or main effects were detected, Tukey’s honestly significant difference (HSD) test was used to compare group means. Asterisks in figures denote significant differences between treatment groups: p < 0.05 (*), p < 0.01 (**), and p < 0.001 (***). For mycelial colonization data, fruiting body yield parameters, and nutritional composition analyses, a one-way analysis of variance (ANOVA) was conducted to evaluate the effects of different treatments. Duncan’s multiple range test was used for post hoc comparisons among means at a significance level of p ≤ 0.05. Results are expressed as means ± standard error, and means within the same row followed by different letters indicate statistically significant differences.

3. Results

3.1. Mycelial Growth Response to Varying Djulis Stem Concentrations in Petri Dish Experiments

3.1.1. Initial Screening (0–100% Djulis Replacement)

Initial screening experiments evaluated the impact of replacing sawdust with djulis stems at high incorporation levels (25%, 50%, 75%, and 100%, designated as DR25, DR50, DR75, and DR100, respectively) on mycelial growth of both P. sajor-caju strains (Figure 1). Pure sawdust (0% djulis stems, CK) served as the control substrate, with all treatments incubated at 28 °C.

During the initial cultivation period (days 1–6), strain TC showed comparable growth rates across most djulis treatments (DR25, DR50, DR75) and the control (CK), with only DR100 exhibiting noticeably slower development. However, by day 12, a clear inhibitory effect emerged for high djulis concentrations, with DR75 and DR100 substrates showing significantly reduced growth rates compared to the control and lower incorporation levels. Strain PT followed a similar pattern, maintaining equivalent growth across treatments during early stages but experiencing marked growth reduction on DR75 and DR100 substrates after day 12.

Analysis revealed that control (CK) and DR25 substrates supported superior mycelial development, achieving colony diameters exceeding 8 cm by day 10. In contrast, high djulis incorporation levels severely compromised growth performance. By day 10, strain TC reached only 4.53 cm diameter on DR100 and 4.85 cm on DR75, while strain PT achieved slightly better performance with 4.87 cm and 5.61 cm diameters, respectively. These results clearly demonstrated that djulis stem incorporation above 50% adversely affected P. sajor-caju mycelial growth, necessitating evaluation of lower incorporation ratios.

3.1.2. Refined Range Test (0–30% Djulis Replacement)

Following the identification of growth inhibition at high djulis concentrations, a refined experiment examined lower incorporation ratios (10%, 15%, 20%, 25%, and 30%, designated as DR10, DR15, DR20, DR25, and DR30) to determine the optimal djulis stem concentration for P. sajor-caju cultivation (Figure 1).

For strain TC, mycelial growth remained statistically equivalent across all djulis treatments throughout the 12-day observation period, indicating good tolerance to djulis incorporation within this concentration range. Strain PT demonstrated similar growth consistency among treatments, with only DR10 showing marginally slower development compared to other ratios. These findings confirmed that P. sajor-caju can successfully colonize substrates containing djulis stems at ratios of 15–25% without significant growth impairment.

Comparative analysis of strains TC and PT at day 8 revealed superior performance of strain PT across multiple djulis treatments (Figure 2). Specifically, strain PT exhibited faster growth rates than TC at DR15, DR20, and DR25 ratios, demonstrating better adaptation to djulis-containing substrates. Based on this strain-specific response and the promising growth performance, strain PT was selected for subsequent large-scale cultivation trials evaluating mycelial colonization, fruiting body production, and nutritional quality using substrate treatments DR15, DR20, and DR25.

3.2. Commercial-Scale Mycelial Colonization in Mushroom Grow Bags

Scale-up experiments using commercial-sized grow bags provided critical insights into the practical applicability of djulis stem incorporation under production conditions (

Table 1. Mycelial growth progression of P. sajor-caju strain PT in mushroom grow bags with different djulis stem ratios (mean ± SEM, n = 24 replicates). Duncan’s multiple range test was performed using SPSS 27.0 for post hoc comparisons at p ≤ 0.05. Means within the same row followed by different letters indicate significant differences.

Substrate	Mycelium Growth Progression (cm)
	DAY7	DAY14	DAY21	DAY28	DAY35
M	4.61 ± 0.22 b	15.46 ± 0.20 a	20.00 ± 0.00 a	20.00 ± 0.00 a	20.00 ± 0.00 a
CK	3.18 ± 0.10 c	8.86 ± 0.14 e	13.70 ± 0.21 c	18.29 ± 0.14 b	20.00 ± 0.00 a
DR15	4.15 ± 0.13 b	11.00 ± 0.18 c	16.70 ± 0.14 b	20.00 ± 0.00 a	20.00 ± 0.00 a
DR20	4.23 ± 0.15 b	10.08 ± 0.18 d	17.04 ± 0.14 b	20.00 ± 0.00 a	20.00 ± 0.00 a
DR25	6.83 ± 0.16 a	14.69 ± 0.12 b	20.00 ± 0.00 a	20.00 ± 0.00 a	20.00 ± 0.00 a

). Mycelial colonization rates varied significantly among treatments, with the commercial sawdust control (M) achieving the fastest substrate colonization in 16 days at a rate of 1.13 cm/day. Among the djulis treatments, DR25 demonstrated superior performance, completing full bag colonization in 20 days with a growth rate of 1.0 cm/day, closely approaching the commercial standard. Moderate djulis incorporation (DR20) required 27 days for complete colonization at 0.74 cm/day, while lower incorporation levels showed slower development. Notably, both DR15 and the sawdust control (CK) exhibited the slowest colonization rates, requiring 28 and 30 days, respectively, with growth rates of 0.71 and 0.67 cm/day.

These results revealed an optimal djulis incorporation range for commercial-scale production, with 25% replacement demonstrating the best balance between substrate utilization and colonization efficiency. The superior performance of DR25 over the sawdust control (CK) suggested potential synergistic effects of djulis stem incorporation on mycelial development under production conditions.

3.3. Fruiting Body Productivity and Biological Efficiency Across Harvest Cycles

Fruiting body production was evaluated across three successive harvest flushes to assess the long-term productivity of djulis-amended substrates (

Table 2. The yield of fruiting bodies and biological efficiency of P. sajor-caju strain PT growing on mushroom grow bags with different djulis stem ratios (mean ± SEM, n = 24 replicates). Duncan’s multiple range test was performed using SPSS 27.0 for post hoc comparisons at p ≤ 0.05. Means within the same row followed by different letters indicate significant differences. Biological Efficiency (%) = (Fresh weight of harvested mushrooms/Dry weight of substrate at inoculation) × 100. The market-bought grow bags (M) contained substrate 0.8 ± 0.2 kg (water content 60%), whereas the grow bags of CK, DR15, DR20, and DR25 contained substrate of 1.2 kg (water content 65%).

Substrate	1st Flush	2nd Flush	3rd Flush	Total Yield	Biological Efficiency
	(g/Bag)	(g/Bag)	(g/Bag)	(g/Bag)	(%)
CK	85.10 ± 8.87 ab	92.46 ± 3.13 c	108.50 ± 17.79 ab	286.10 ± 3.7 a	71.53%
DR15	68.40 ± 14.26 ab	157.20 ± 7.163 a	94.30 ± 7.759 b	319.90 ± 14.5 a	76.17%
DR20	73.30 ± 4.57 ab	108.32 ± 4.79 bc	124.90 ± 10.23 a	306.50 ± 17.5 ab	72.98%
DR25	91.50 ± 11.6 a	115.28 ± 3.79 b	94.50 ± 3.87 b	301.33 ± 11.9 a	71.75%
M	52.50 ± 3.63 b	107.90 ± 4.79 bc	85.40 ± 2.72 b	245.78 ± 27.7 b	76.80%

). Total yields among experimental treatments (CK, DR15, DR20, DR25) showed no significant differences, ranging from 286.10 ± 3.7 g to 319.90 ± 14.5 g per bag, demonstrating that djulis incorporation up to 25% maintained productive capacity. Biological efficiency calculations revealed important distinctions between treatments. The commercial control (M) achieved 76.80% biological efficiency, while among djulis treatments, DR15 demonstrated the highest efficiency at 76.17%, followed by DR20 (72.98%) and DR25 (71.75%). The sawdust control (CK) showed intermediate efficiency at 71.53%. It should be noted that biological efficiency comparisons account for differences in substrate mass between commercial bags (0.8 ± 0.2 kg at 60% moisture) and experimental bags (1.2 kg at 65% moisture). These findings confirm the commercial viability of djulis stem incorporation, with 15–25% replacement ratios maintaining both yield and efficiency comparable to conventional sawdust substrates.

3.4. Impact of Djulis Incorporation on Mushroom Nutritional and Bioactive Profiles

Nutritional composition analysis (

Table 3. Nutritional and bioactive composition analysis of fruiting bodies produced under different treatments (mean ± SEM, n = 5 replicates). All nutritional and amino acid values are expressed per 100 g fresh weight (fw). Total phenolics and flavonoids are expressed in mg per g dry weight (dw). Antioxidant activities are presented as unitless percentages or optical density. Duncan’s multiple range test was performed using SPSS 27.0 for post hoc comparisons at p ≤ 0.05. Means within the same row followed by different letters indicate significant differences.

Component	Treatment
	CK	DR15	DR20	DR25	M
Moisture (g/100 g fw)	90.4 ± 0.52 a	90.1 ± 0.49 a	89.1 ± 0.31 a	89.7 ± 0.19 a	90.3 ± 0.27 a
Crude fat (g/100 g fw)	2.10 ± 0.57 a	1.77 ± 0.91 a	1.44 ± 0.61 a	1.48 ± 0.65 a	1.98 ± 0.14 a
Crude ash (g/100 g fw)	0.66 ± 0.04 a	0.67 ± 0.03 a	0.74 ± 0.02 a	0.73 ± 0.01 a	0.68 ± 0.02 a
Crude protein (g/100 g fw)	4.14 ± 0.22 a	3.90 ± 0.36 a	4.32 ± 0.18 a	3.92 ± 0.12 a	3.98 ± 0.21 a
Total carbohydrates (g/100 g fw)	2.70 ± 0.36 c	3.56 ± 0.15 abc	4.44 ± 0.17 a	4.20 ± 0.24 ab	3.06 ± 0.10 c
Total free amino acids (TFAA, g/100 g fw)	0.50 ± 0.07 ab	0.43 ± 0.09 b	0.63 ± 0.10 a	0.69 ± 0.11 a	0.64 ± 0.12 a
Total branched-chain amino acids (BCAA, g/100 g fw)	0.18 ± 0.01 b	0.21 ± 0.05 ab	0.18 ± 0.02 b	0.21 ± 0.01 ab	0.22 ± 0.01 a
Lysine (g/100 g fw)	0.06 ± 0.00 a	0.07 ± 0.009 a	0.06 ± 0.002 a	0.07 ± 0.009 a	0.06 ± 0.008 a
Total phenolics (mg/g dw)	25.57 ± 0.33 c	27.78 ± 0.51 b	22.31 ± 0.67 e	23.41 ± 0.12 d	32.54 ± 0.98 a
Total flavonoids (mg/g dw)	4.80 ± 0.06 a	3.20 ± 0.05 d	4.10 ± 0.05 b	3.40 ± 0.09 c	4.10 ± 0.04 b
Free radical scavenging activity (% DPPH)	75.00 ± 3.31 a	72.03 ± 10.62 ab	57.10 ± 10.1 b	68.80 ± 7.90 ab	59.30 ± 5.29 b
Reducing power (OD700nm)	0.46 ± 0.03 bc	0.43 ± 0.017 c	0.41 ± 0.02 c	0.56 ± 0.03 a	0.52 ± 0.04 ab

) showed that djulis stem incorporation did not lead to significant differences among treatments in terms of moisture, crude fat, crude ash, and crude protein; whereas higher total carbohydrates were observed in DR20 and DR25 compared to CK and M. Similarly, bioactive profiles of treatments showed comparable results, except the total free amino acids (TFAA) and reducing power. TFAA increased with djulis concentration, reaching 0.69 g/100 g in DR25 compared to 0.50 g/100 g in the control (CK), and reducing power (OD_700nm) increased to 0.56 compared to 0.46. Moreover, antioxidant compound analysis revealed mixed responses to djulis incorporation, and total phenolic and flavonoid contents showed no consistent pattern across treatments.

4. Discussion

The superior performance of strain PT over strain TC in djulis-containing substrates (Figure 2) suggested genetic variation in substrate adaptation capabilities among P. sajor-caju isolates. This strain-specific response highlights the importance of careful strain selection when developing alternative substrate formulations, as different genetic backgrounds may exhibit varying tolerance to components such as the saponins and secondary metabolites present in djulis stems.

The clear inhibitory effect observed at djulis incorporation levels above 50% (Figure 1A,B) provides critical insights into the physiological limits of P. sajor-caju substrate tolerance. This threshold effect likely resulted from the accumulation of inhibitory compounds, particularly saponins characteristic of Chenopodium species, which are known to disrupt cellular membrane integrity and metabolic processes at high concentrations [14]. Although a soaking pre-treatment was applied to reduce water-soluble saponins, the residual presence at higher djulis ratios could still impair mycelial growth. However, since saponin concentrations were not directly measured in this study, this hypothesis remains speculative. Future research should incorporate quantitative saponin analysis to validate the proposed inhibitory mechanism and refine substrate optimization strategies.

The water requirements for djulis pre-treatment represent both a practical consideration and an opportunity for process optimization. While the 8–12 h soaking process requires 2–3 L water per kg of djulis stems, this must be evaluated within the broader context of sustainable mushroom production. The water footprint of djulis processing should be compared against the environmental costs of sawdust production, including deforestation, transportation, and processing operations. Moreover, the saponin-rich processing water could potentially be valorized as a natural surfactant for agricultural applications or processed for saponin extraction, creating additional economic value while implementing water recycling systems. Future technological developments could further reduce water requirements through batch processing with recirculation systems (such as those demonstrated in coffee wet processing systems), steam treatment alternatives, or enzymatic pre-treatment methods to minimize both processing time and water consumption.

The performance of djulis stems in P. sajor-caju cultivation can be contextualized within the broader landscape of agricultural waste utilization for oyster mushroom production. Similar to our findings with 15–25% djulis incorporation, other studies have demonstrated successful partial replacement strategies with various agricultural residues. Corn cob and rice husk substrates achieved biological efficiencies of 78.4% and 71.6%, respectively, when used as sole substrates for P. ostreatus cultivation, which closely parallel our DR15 treatment achieving 76.17% biological efficiency [22]. Similarly, wastepaper supplemented with wheat bran yielded biological efficiency of 71.2% in P. ostreatus production, demonstrating that moderate incorporation levels consistently optimize performance across diverse agricultural waste combinations [23]. These results suggest that the 15–30% incorporation range represents a universal optimization strategy across diverse agricultural waste substrates, regardless of the specific residue type.

The superior performance of certain agricultural residues over traditional sawdust substrates, as observed with our DR25 treatment exhibiting faster mycelial growth (1.0 cm/day) than the sawdust control (0.67 cm/day), has been documented with other waste materials. P. ostreatus grown on rice straw achieved biological efficiency of 85.2% compared to 76.8% on sawdust controls, while also demonstrating faster colonization rates [24]. Similarly, P. pulmonarius cultivation on rice straw and banana leaves achieved biological efficiencies of 68.3% and 71.2%, respectively, both superior to conventional sawdust substrates [25]. Cotton waste substrates also demonstrated enhanced performance, with P. ostreatus achieving 74.5% biological efficiency compared to 65.8% on sawdust controls [26]. This phenomenon likely results from the diverse nutrient profiles and mineral compositions of agricultural waste providing complementary growth factors absent in pure sawdust formulations.

Interestingly, the superior performance of DR25 over the sawdust control (CK) in commercial-scale bags (Table 1) suggested potential synergistic effects between djulis components and sawdust at optimal ratios. This enhancement may result from djulis stems providing additional nutrients, such as minerals or nitrogen compounds, that complement the carbon-rich sawdust matrix. The lignocellulosic composition of djulis stems, while similar to other agricultural residues successfully used for mushroom cultivation [4], appears to offer unique benefits when properly balanced with traditional substrate components.

The maintenance of yield and biological efficiency comparable to commercial standards (Table 2)demonstrated the practical feasibility of djulis incorporation for industrial mushroom production. The 15–25% incorporation range identified in this study provides producers with flexibility to optimize substrate formulations based on djulis availability and cost considerations. The biological efficiency of 76.17% achieved with DR15 rivals that of commercial substrates (76.80%), indicating that partial sawdust replacement by djulis stems would not compromise production economics.

The three-flush harvest cycle revealed stable productivity across multiple cropping cycles (Table 2), addressing concerns about substrate longevity and sustained nutrient availability. This consistency is crucial for commercial operations where predictable yields are essential for economic planning. The successful substrate colonization observed across djulis treatments reflects the robust ligninolytic enzyme systems characteristic of Pleurotus species. These fungi produce a comprehensive array of lignin-degrading enzymes, including laccase, manganese peroxidase, and lignin peroxidase, which facilitate the breakdown of complex lignocellulosic structures present in both sawdust and djulis stems [27]. The enhanced mycelial growth observed with DR25 compared to pure sawdust suggests that djulis stems may serve as effective inducers for ligninolytic enzyme production, potentially due to their unique lignin composition and the presence of phenolic compounds that can act as enzyme mediators. Agricultural residues with varied lignin compositions and structures can stimulate differential enzyme expression patterns, with wheat straw and corn stalks inducing higher laccase activity (up to 150 U/L) while hardwood sawdust promotes manganese peroxidase production (up to 85 U/L), potentially explaining the substrate-specific growth responses observed in our study [28]. The maintenance of consistent productivity across three harvest flushes indicates sustained enzyme activity throughout the cultivation cycle, confirming the compatibility of djulis lignocellulosic components with P. sajor-caju’s enzymatic machinery. The ability to maintain production levels while incorporating agricultural waste aligns with circular economy principles and provides a valuable outlet for the approximately 200 hectares of djulis cultivation residues generated annually in Taiwan.

Previous research has shown that alternative substrates can alter certain nutritional parameters, including amino acid content, phenolic compounds and antioxidant activities in the fruiting bodies of various mushrooms [15,16,17]. However, discordant with those studies, the nutritional and bioactive profile analysis (Table 3) revealed that replacing culture substrate with djulis stem did not result in major changes in nutritional values. Exceptions were some enhancements in reducing power and total free amino acid content was observed when djulis ratio increased to 25%. The lack of major differences in amino acid profiles and DPPH scavenging activity across all djulis incorporation levels demonstrates that substrate modifications do not compromise mushroom quality parameters. This finding is particularly significant for commercial producers, as it confirms that djulis stem incorporation maintains the nutritional and functional properties that consumers expect, while simultaneously enabling sustainable waste utilization without sacrificing product marketability or nutritional value.

The successful incorporation of djulis stems addresses multiple sustainability challenges simultaneously. By reducing sawdust requirements by 15–25%, this approach could decrease pressure on forest resources while providing economic value for djulis cultivation waste that would otherwise require disposal. Based on Taiwan’s annual sawdust consumption of 226,000 metric tons by the mushroom industry [29], a 15–25% substitution with djulis stems would reduce sawdust demand by 33,900 to 56,500 metric tons annually. Furthermore, the economic feasibility of djulis stems as a sawdust substitute under Taiwan market conditions was assessed as follows.

At present, djulis stems lack an established commercial market and are typically left in fields or incorporated into soil after harvest. This situation is similar to that of its close relative quinoa (Chenopodium quinoa), where major producing countries such as Peru and Bolivia also do not utilize stem byproducts commercially. In Taiwan, farmers have access to small-scale machinery for harvesting djulis stems, but the lignified nature of the fibers necessitates chopping, which adds to processing costs. From the perspective of economic feasibility, a comprehensive cost build-up under Taiwan’s agricultural conditions yields an estimated production cost of USD 45–75/ton on a dry-weight equivalent basis, considering harvesting, chopping, drying and conditioning, storage, and local transportation. With a 20% profit margin, the realistic market price is projected at USD 60–90/ton delivered. Rice straw, which has similar harvesting and residue-handling requirements, is commonly used as a mushroom substrate and thus serves as a practical pricing benchmark, with a current market value of USD 130.7/ton on a dry-weight equivalent basis. At a market price of USD 60–90/ton, djulis stems have a 58–75% cost advantage compared with sawdust priced at USD 215.7–238.6/ton on a dry-weight basis. Assuming a standard cultivation bag of approximately 1.0 kg at 60% moisture content (used here for calculation purposes), the production of 10,000 P. sajor-caju grow bags requires about 4000 kg of dry sawdust, equivalent to a substrate cost of USD 862.8–954.4. Replacing 15% of sawdust with djulis stems reduces substrate costs by USD 76–107, corresponding to savings of approximately 9–11%, whereas a 25% substitution lowers costs by USD 127–179, equivalent to savings of about 15–19% compared with the all-sawdust baseline. Even if the price of djulis stems increases toward the upper bound of USD 90/ton due to supply chain development, transportation logistics, or seasonal constraints, the material would still remain approximately 60% cheaper than sawdust, demonstrating economic viability under fluctuating market conditions.

Food safety is a critical consideration when using agricultural by-products as substrates for edible mushroom cultivation, particularly with respect to potential contamination by heavy metals and other elements. The accumulation of such elements in fruiting bodies usually reflects contamination from substrate materials or the growing environment. To ensure safe commercial use, it is recommended that djulis stems and other substrate components be sourced from trusted suppliers and prepared according to standard industry practices to minimize contamination risks. Systematic testing for both heavy metals and nutritionally relevant minerals is also necessary to meet regulatory standards and protect consumer health.

5. Conclusions

This study successfully demonstrated the feasibility of using djulis (Chenopodium formosanum) stems as a partial replacement for sawdust in P. sajor-caju cultivation, representing a novel approach to sustainable mushroom production and agricultural waste utilization. The identification of 15–25% djulis incorporation as the optimal range provides practical guidance for commercial implementation, with 25% replacement achieving superior mycelial growth rates and 15% replacement maximizing biological efficiency. The maintenance of yield, biological efficiency, and nutritional quality comparable to commercial standards confirms the commercial viability of this substrate modification. Enhanced total free amino acid content and selective improvements in antioxidant properties suggest potential value-added benefits beyond sustainability considerations. These findings contribute to addressing Taiwan’s mushroom industry challenges related to sawdust shortages and environmental sustainability while providing economic value for djulis cultivation residues. The successful incorporation of up to 25% djulis stems without compromising production parameters offers a practical solution for reducing forest resource dependence while maintaining product quality. Future research should prioritize three key areas: optimizing water-efficient pre-treatment methods with valorization of saponin-rich wastewater, conducting comprehensive heavy metal analysis of both substrate materials and fruiting bodies for food safety compliance and regulatory approval, and performing detailed economic evaluation including cost–benefit analysis and market pricing assessment. Broader applications across different mushroom species will be needed to fully assess and maximize the impact of this sustainable cultivation approach.