Introduction

Biofilms are microbial communities attached to various substrates (e.g. plants, wood piles, rocks) in submerged surfaces and are composed of autotrophic and heterotrophic microorganisms including algae, bacteria, fungi, and protozoa [1, 2]. The primary producers in biofilms, i.e., algae, therein play a fundamental role in primary production as basal resources for consumers in stream food webs [3,4,5]. Algae, in particular Bacillariophyta, usually containing abundant polyunsaturated fatty acid (PUFA), provide high-quality food for herbivores compared to allochthonous sources, e.g. terrestrial litter in river ecosystems [6,7,8]. The indispensability of PUFAs for growth, neural development, and reproduction in aquatic invertebrate consumers has been well-established [9, 10]. Nevertheless, PUFA are either incapable of being synthesized by animals or are synthesized at a negligible rate, which often choose direct sources from high nutritional quality basal resources—algae in biofilms [11, 12].

Algae are sensitive to changes in ambient environmental variables [13, 14]. As riparian canopy has been opened led by deforestation for agricultural practices, the amount of light that reaches the stream channel has been altered, consequently influencing the composition of benthic algae [14]. The composition of stream periphytic algal communities is commonly dominated by Bacillariophyta, Chlorophyta, and Cyanobacteria [15]. Cyanobacteria, which lack eicosapentaenoic acid (EPA, C20:5n3) and docosahexaenoic acid (DHA, C22:6n3), are abundant in α-linolenic acid (ALA; C18:3n-3), a precursor fatty acid that can be synthesized into EPA and DHA [16, 17]. Chlorophyta, usually required high light levels [18], are considered to be of medium dietary quality, containing short-chain PUFA such as ALA and linoleic acids (LIN, C18:2n6) but relatively deficient in EPA [17]. Bacillariophyta favour lower light than Chlorophyta [19], and are considered high-quality food sources for aquatic invertebrates due to their content of long-chain PUFA (LC-PUFA, including ARA, EPA, and DHA) [20]. Hence, shifts in community composition result in changes of fatty acid profiles in biofilms, subsequently influencing the availability of n-3 and n-6 PUFA [1, 21, 22].

Transcriptomic analysis could provide a molecular perspective on fatty acid synthesis. It has observed downregulation of genes in the photosynthetic apparatus that encode key proteins such as light-harvesting complexes (LHC) components under reduced light conditions [23]. Downregulation in fatty acid synthesis such as acetyl-CoA carboxylase (ACCase) and fatty acid synthase (FAS) could lead to a reduction in overall lipid content [24]. Moreover, key enzymes in the PUFA biosynthesis pathway, such as desaturases and elongases, exhibit altered expression levels under different light intensity [23, 25].

In pristine forested streams, biofilms are regulated directly from the bottom-up by light and nutrient availability, while in deforested streams, light limitation on biofilms is expected to be relaxed [26]. As human activities alter light regimes in many streams such as deforestation (increased light) [27], the alteration of biofilms nutritional quality is likely to cascade upward and influence organisms in higher trophic levels [28, 29]. To counteract these changes and promote instream ecosystem recovery, riparian revegetation is usually undertaken by focusing on the crucial functions that riparian zones provide for aquatic ecosystems [30]. While studies have conducted manipulative experiments in the field to examine the effects of abiotic factors on FA profiles [28, 29, 31], others have focused on exploring the transcriptional responses of specific algal species to different light intensities [23, 32, 33]. However, it remains unclear how the fatty acid composition is molecularly adjusted, coupled with community succession, with riparian revegetation.

In this study, we hypothesized that (i) shading, change in light intensity, would lead to alteration of PUFA in biofilms FA profiles and (ii) the alteration of FA profiles is attributable not only to the succession of algal community but also to the changes in differentially expressed genes in metabolic pathways. We intended to reveal change of nutritional quality of basal resources in food webs with multidimensional mechanisms at algal community and molecular levels, i.e., lipid metabolism and energy metabolism. Our study sheds light on the changes of the nutritional quality in biofilms with riparian deforestation.

Materials and Methods

Manipulative Experiment in Mesocosm

The manipulative experiment was performed in a greenhouse in Wuhan Botanical Garden, Chinese Academy of Sciences in 2021 for a period of 35 days (from June 18 to July 23). Six cylindrical polyethylene buckets (volume: 1000 L, diameter: 1.2 m, height: 0.8 m) were used to simulate stream ecosystems. A pump was installed in each mesocosm to allow the flowing of water (Fig. 1a). The bottom of the mesocosms was covered with a layer of cobbles. To closely simulate natural river biofilms, we collected cobbles already colonized by biofilms from the Chuka River (115°19′10″E, 31°5′50″N) in Huanggang City, Hubei Province (Fig. 1b), an area with minimal human disturbance. We collected the cobbles with colonized biofilms from the Chuka River between 10:00 and 14:00 in a day, during which the water temperature in unshaded areas ranged from 28.4 to 31.6 °C. These cobbles were subsequently placed in the mesocosms for 1 week, allowing the biofilms to acclimate to their new habitats before the experiments started.

Fig. 1
figure 1

a The schematic diagram of experiment showing the “Ambient” (ambient light intensity) and the “Shaded” (added shaded nets) treatments and b the habitat in streams for biofilms in the mesocosm experiment

Full size image

Six mesocosms were divided into two groups. One group was covered with shade cloth to simulate riparian canopy conditions (shading groups), while the other group was exposed to ambient light and served as the control (ambient groups). Initial properties of the water were total dissolved nitrogen (TDN) of 1.34 ± 0.04 mg/L and total dissolved phosphorus (TDP) of 0.02 ± 0.00 mg/L. At noon, an illuminometer measured light intensity, revealing 141,000 ± 29,813 lx under ambient light and 827 ± 113 lx in shaded conditions. It should be noted that this significant reduction in light intensity is not uniform throughout the day, as light intensity naturally fluctuates with the time of day.

Sample Collection

Physical and Chemical Variables

In mesocosm, water temperature (WT), pH, dissolved oxygen (DO), electrical conductivity (EC), and ammonium (NH4+) were measured using a YSI EXO3 multi-parameter water quality meter (YSI Inc., Yellow Springs, OH, USA). The sensors were calibrated before measurements were taken. The samples of the surface water at each mesocosm in triplicates were collected and filtered using cellulose nitrate membrane filters (Whatman, 0.45 µm pore size) for the analysis of total dissolved nutrient, specifically total dissolved nitrogen (TDN), total carbon (TC), and total dissolved phosphorus (TDP). Water samples were measured and collected in pre- and post-treatment, as well as once a week throughout the shading treatment. Samples for analysis of nutrients were stored on ice in a cooler, and then frozen before analysis.

Biofilms

Five different cobbles were randomly collected, and an area with a 10-cm diameter on each rock was brushed. The biofilms from five rocks was rinsed with distilled water into one container forming a composite sample of around 50 mL as one replicate. The 5-mL samples were taken from 50 mL and preserved in 4% formaldehyde. The remaining samples (45 mL) were stored at − 80 ℃ and used for fatty acid analysis and transcriptome sequencing. Biofilms were filtered onto pre-ashed glass fibre filters (0.7 µm; Whatman GF/F filters) and stored at − 20 ℃ for FA analysis. Three replicates were made at each treatment.

Laboratory Analyses

Physical and Chemical Variables

Both TC and TDN were measured using a TOC/TN analyzer equipped with different modules for measuring these components (Elementar Corporation). Total dissolved phosphorus (TDP) was measured by inductively coupled plasma atomic emission spectrometer (ICP-OES) (Thermo Fisher, X Series 2, USA).

Biofilms Collected for Identification and FA Analysis

The biofilms containing samples (5 mL) for algal identification and counting were preserved in plastic bottles with Lugol’s solution. The samples were thoroughly mixed, and a 1-mL subsample was taken for identification and enumeration, conducted using a Sedgewick Rafter cell under a microscope (Olympus BX51, Olympus Corporation, Tokyo, Japan) with 400 × magnification [34]. Enumeration ceased when 40 squares were counted under the microscope. The slide was then scanned to 200 squares at 200 × magnification to identify any algae not previously sighted [35]. Taxa were combined into categories (Bacillariophyta, Cyanobacteria, and Chlorophyta) for analysis, and the relative proportions of each category were calculated based on the number of cells.

Prior to fatty acid analyses, samples were freeze dried for 48 h and homogenized. The three replicate samples from the same mesocosm were pooled before analyses to guarantee a sufficient amount of sample material. Lipids were extracted and methylated from the pooled samples (400 mg) using the following method [36]: A 15-mL mixture of 0.2 M KOH:MeOH (1:1) was maintained at 50 °C for 60 min, with the tubes vortexed every 10 min to lyse and saponify the cells. At the end of this process, 10 mL of n-hexane were added to extract the fatty acid methyl esters (FAMEs) from the acidic aqueous phase into the organic phase, and the upper organic layer was transferred. To ensure thorough extraction of the FAMEs, this process was repeated twice. To enhance the detection rate of FAMEs, the organic solvent was dried and concentrated under N2, and an internal standard solution (methyl undecanoate) was added.

FAMEs were analyzed using a gas chromatograph equipped with a mass detector (GC–MS, Agilent 7890B + 5975C, USA), which is equipped with a temperature programmable injector and an autosampler. Temperatures of the interface and ion source were 270 °C and 230 °C, respectively. Agilent J&W HP-88 column was used with the following temperature program: 100 ℃ was maintained for 5 min, then to 170 ℃ at a rate of 10 ℃/min for 8 min, then to 200 ℃ at a rate of 1 ℃/min for 20 min, and finally heated to 230 ℃ at a rate of 8 ℃/min for 10 min. Total program time was 84.75 min and solvent cut time 10 min. Helium was used as carrier gas, constant pressure mode is at 32 psi, injection volume is 1 µL, and split flow is 5:1. The identity of the FAMEs was determined by comparing the retention times of unknown sample peaks with those of known standards, including a 37-component FAME mix (Supelco No. 47885-U) and Bacterial Acid Methyl Ester Mix (Supelco No. 47080-U). Fatty acid concentrations were calculated using free fatty acid of C11:0 as internal standards. FA results were expressed as percentages relative to total FA (FA %).

Transcriptome Data Processing

RNA Extraction, Library Construction, and Sequencing

Total RNAs were extracted from the biofilm samples with four replicates using TRIzol reagent according the manufacturer’s instructions (Invitrogen). RNA quality was determined using 2100 Bioanalyzer (Agilent) and quantified using the ND-2000 (NanoDrop Technologies). High-quality RNA sample (OD260/280 ≥ 1.8, OD260/230 ≥ 1.0, RIN ≥ 6.5, 28S: 18S ≥ 1.0, > 1 µg) was used to construct sequencing library.

RNA purification, reverse transcription, library construction, and sequencing were conducted by Shanghai Majorbio Bio-Pharm Biotechnology Co., Ltd. (Shanghai, China). For the biofilms RNA-seq transcriptome library, 1 µg of total RNA was employed. Initially, messenger RNA was isolated through the poly(A) selection method using oligo (dT) beads and subsequently fragmented using a fragmentation buffer. Next, double-stranded cDNA was synthesized using the SuperScript double-stranded cDNA synthesis kit (Invitrogen, CA) with random hexamer primers (Illumina). Libraries were size selected for cDNA target fragments of 200–300 bp on 2% Low Range Ultra Agarose followed by PCR amplified using Phusion DNA polymerase for 15 PCR cycles. The PCR protocol included an initial denaturation step at 98 °C for 30 s, followed by 15 cycles of denaturation at 98 °C for 10 s, annealing at 60 °C for 15 s, and extension at 72 °C for 30 s. A final extension was performed at 72 °C for 5 min. After quantified by TBS380, paired-end libraries were sequenced by Illumina NovaSeq 6000 sequencing.

Transcriptome Assembly and Functional Annotation

For quality control, the raw reads were processed using SOAPnuke software (v2.1.0) to obtain high-quality clean reads. This involved steps such as image recognition, decontamination, removal of joints, removal of adaptor sequences, ambiguous reads (“N”), and low-quality reads (reads with more than 10% ambiguous bases or with bases below a quality score of 20) [37, 38]. Then, clean data from the biofilms (“shaded groups” and “ambient groups”) were used to do de novo assembly with Trinity software [39].

Gene function annotation utilized BLASTx with an E-value threshold of 1.0 × 10−5 against the following databases: NR (NCBI nonredundant protein sequences), KEGG (Kyoto Encyclopedia of Genes and Genomes), COG (Clusters of Orthologous Groups of proteins), Swiss-Prot, Pfam, and GO (Gene Ontology).

Differential Expression Genes (DEGs) Analysis and Functional Enrichment

To identify differential expression genes (DEGs) between two groups, we quantified transcript abundances using the transcripts per million reads (TPM) method and RSEM [38]. Differential expression analysis of two groups was conducted using the DEGseq R package. DEGs were identified based on a Poisson distribution with a false discovery rate (FDR) ≤ 0.001 and a fold change ≥ 2 (log2 ratio ≥ 1) between the two groups. In addition, functional enrichment analysis including GO and KEGG was performed to identify which DEGs were significantly enriched in GO terms at Bonferroni-corrected p-value ≤ 0.05 compared with the transcriptome background.

Data Analysis

All data were obtained by using at least three biological samples to ensure the fidelity of the results. The following FA were used for data analyses: LIN, ALA, ARA, EPA, DHA, the sum of saturated fatty acids (SAFA), the sum of monounsaturated fatty acids (MUFA), the sum of polyunsaturated fatty acids (PUFA), the sum of long-chain polyunsaturated fatty acids (LC-PUFA, typically containing 20 or more carbon atoms), and the sum of C18 PUFA. The independent samples T test was used to test the effects of shading on biofilms’ FA profiles and algal community structure after 5 weeks treatment. The statistical analyses were conducted in the statistical software IBM SPSS Statistics 22.

In transcriptome data, the DEGs output was further enriched through GO and KEGG annotation, with enrichment measured by rich factor, -log10 (p-value), and gene number. The results of gene set enrichment analysis (GSEA) are typically reported as enrichment scores, which indicate the level of enrichment of a specific gene set within the ranked list. Significantly enriched gene sets were selected based on the following: a significance level of p < 0.05 and an adjusted p-value (FDR) threshold of < 0.25. Statistical analyses were conducted using the statistical software R version 4.0.0 (R Core Team, 2020).

Result

Physicochemical Variables in the Mesocosms

At the end of the experiment, water temperatures (p = 0.100) and the concentration of dissolved nutrients, including total dissolved nitrogen (TDN) and total dissolved phosphorus (TDP), did not show significant changes in the shaded groups compared to the ambient groups (p = 0.381 and p = 0.944, respectively). Dissolved oxygen (DO) and pH exhibited significant decreases (p = 0.025 and p = 0.006, respectively), while electrical conductivity (EC) was significantly higher (p < 0.001) in the shaded groups. A relative increase in total carbon (TC) content was observed compared to the ambient groups (p = 0.101) (Table 1).

Table 1 The physical and chemical parameters in waters (means ± SD) in the mesocosms with different treatments
Full size table

Algal Community Analysis

Analysis of the abundance of three primary algal groups—Bacillariophyta, Cyanobacteria, and Chlorophyta—showed that shading altered the periphyton algal community structure. Initially, 26 genera from three phyla of algae were identified. After 5 weeks of shading treatment, the number of genera within the phyla Bacillariophyta, Cyanobacteria, and Chlorophyta decreased in both shaded and ambient groups (Table 2). The average proportion of Bacillariophyta was significantly higher in the shaded groups (66.15%) compared to the ambient groups (52.66%) (p = 0.027). Conversely, Cyanobacteria were significantly lower in the shaded groups (1.87%) compared to the ambient groups (5.62%) (p = 0.035), while Chlorophyta showed borderline significance (p = 0.052) with a lower proportion in the shaded groups (Fig. 2, Table 2).

Table 2 Mean abundance of algal taxa (genera) in biofilms in manipulative experiment
Full size table
Fig. 2
figure 2

The percentage of Bacillariophyta, Cyanobacteria, and Chlorophyta of algal communities in biofilms after 5 weeks of shading treatment. *p < 0.05, **p < 0.01, ***p < 0.001

Full size image

Fatty Acid Profiles of Biofilms

After 5 weeks of shading, total PUFA% and LC-PUFA% including ARA% significantly increased compared to the ambient groups (Fig. 3a, b). However, EPA% and DHA% did not change significantly (p = 0.986 and p = 0.904, respectively) (Fig. 3a). Additionally, SAFA% decreased significantly (p = 0.040), particularly C16:0% (p = 0.001). MUFA showed no significant change (p = 0.326) compared to the ambient groups, while the fatty acid biomarker C16:1n7 indicative of diatoms showed a significant increase (p = 0.041) (Fig. 3b, Table 3).

Fig. 3
figure 3

Fatty acid profiles in biofilms (% of total fatty acids, mean ± SD) in the ambient and shaded groups after 5-week shading treatment in mesocosms. a Five essential fatty acids: LIN, linoleic acid (C18:2n6); ALA, α-linoleic acid (C18:3n3); ARA, arachidonic acid (C20:4n6); EPA, eicosapentaenoic acid (C20:5n3); DHA, docosahexaenoic acid (C22:6n3). b Main fatty acid groups: PUFA, polyunsaturated fatty acids; C18 PUFA, C18 polyunsaturated fatty acids; LC-PUFA, long chain-polyunsaturated fatty acids; MUFA, monounsaturated fatty acid; SAFA, saturated fatty acids. *p < 0.05, **p < 0.01, ***p < 0.001

Full size image
Table 3 Fatty acid compositions (% relative to total FA, mean ± SD) of biofilms after the initial and 5 weeks manipulative experiment in mesocosms. Note: Fatty acids with average proportions < 0.10% was excluded for statistical analysis
Full size table

Transcriptome Analysis

Gene Function Annotation and Classification

The output of transcriptome sequencing of biofilms (shaded and ambient groups) was obtained (Table S1). A total of 832,836 unigenes which can be classified into three major GO categories: biological processes (BP), cell component (CC), and molecular function (MF) (Figure S1a). Genes of the BP group were divided into 25 subcategories with “cellular process” as the largest terms with 426,125 unigenes. The genes in the CC group were divided into 16 subcategories, and “cell part” as the largest terms. In the MF group, the genes were divided into 18 subcategories, where “catalytic activity” was the largest terms.

The metabolic pathway of KEGG was divided into five parts, “cellular processes”, “environmental information processing”, “genetic information processing”, “metabolism”, and “organismal systems” (Fig. S1b). Among these, the pathway represented by the most unigenes was “metabolism” (326,576, 68.80% of the total annotated transcripts). Within the “metabolism” category, carbohydrate metabolism (50,077, 15.33%) was the most represented, followed by energy metabolism (39,708, 12.16%). Our study specifically focuses on lipid metabolism, which ranked fifth at 24,479 unigenes (7.50%).

DEGs and Pathways Enrichment Analysis

Among the 53,908 DEGs, 27,397 genes were found to be significantly upregulated, while 26,511 genes were downregulated in the shaded group compared to the ambient group in response to shading treatment (Figure S2). The KEGG enriched pathways related to fatty acid metabolism, comprising photosynthesis of antenna proteins, photosynthesis, arachidonic acid metabolism, and carbon fixation in photosynthetic organisms, were found to be significant (Fig. 4).

Fig. 4
figure 4

KEGG enrichment analysis was performed for DEGs in the shaded and ambient groups and several pathways related to fatty acids were listed

Full size image

The fatty acid biosynthesis pathway was downregulated, while the pathways for fatty acid elongation and the biosynthesis of unsaturated fatty acids were upregulated (Fig. 5). In the fatty acid biosynthesis pathway, the genes encoding the catalyzing enzymes acyl-ACP desaturase (FAB2, EC: 1.14.19.2) and fatty acyl-ACP thioesterase A (FATA, EC: 3.1.2.14) were significantly downregulated. In the fatty acid elongation pathway, the expression of genes trans-2-enoyl-CoA reductase (MECR, EC: 1.3.1.38) and palmitoyl protein thioesterase (PPT, EC: 3.1.2.22) in mitochondria, as well as 3-ketoacyl-CoA synthase (KCS, EC:2.3.1.199) in endoplasmic reticulum, was upregulated. In the biosynthesis of unsaturated FA pathway, Δ12 desaturase (FAD2, EC:1.14.19.6) actively participates and catalyzes the synthesis towards PUFA. Gene set enrichment analysis (GSEA) revealed significant enrichment in “Map00590” (arachidonic acid metabolism) (NES = 2.64, p < 0.001) and “Map00062” (fatty acid elongation) (NES = 1.51, p < 0.05) (Fig. 6a, b).

Fig. 5
figure 5

The effect of shading on the Calvin cycle, glycolysis, and fatty acid metabolism-related processes in biofilms. Key enzymes are presented as their names in red colours (upregulated) and in blue colours (downregulated). ACSL, long-chain acyl-CoA synthetase; ACP, transacylase; FAB2, acyl-ACP desaturase; FAD2, Δ12 desaturase; FATA, fatty acyl-ACP thioesterase A; KCS, 3-ketoacyl-CoA synthase; MECR, mitochondrial trans-2-enoyl-CoA reductase; PPT, palmitoyl protein thioesterase; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; HK, hexokinase; PK, pyruvate kinase; PDH, pyruvate dehydrogenase; Rubisco, ribulose-bisphosphate carboxylase; 1,3-bis-PG, 1,3-biphosphoglycerate; G-3-P, glyceraldehyde-3-phosphate; G-6-P, glucose-6-phosphate; PEP, phosphoenolpyruvate

Full size image
Fig. 6
figure 6

Gene set enrichment analysis (GSEA) reveals the activation/suppression status of a arachidonic acid metabolism (ARA) and b fatty acid elongation pathways

Full size image

In response to the shading, there was a downregulated key genes encoding the ferredoxin—NADP+ reductase (petH, EC:1.18.1.2) in the photosynthetic electron transport, which regulates NADPH synthesis (Fig. 7). Moreover, 11 out of the 12 genes (except LHCB7) encoding photosynthetic antenna proteins in the light-harvesting complex I (LHCA) and light-harvesting complex II (LHCB) families exhibited upregulation (Table S2). Most components of photosystem I (PSI), except for psaA and psaB, were upregulated. In contrast to PSI, the components associated with PSII, including PsbD, PsbK, and PsbS, were downregulated. Similarly, a substantial downregulated expressional level of three components of F-type H+ transporting ATPase (atpA, atpB, atpD, and atpH, Table S2) was observed. During the carbon fixation pathways, ribulose-1,5-bisphosphate (RuBP) was converted to glyceraldehyde-3-phosphate (G-3-P) by ribulose-bisphosphate carboxylase (Rubisco) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH, EC:1.2.1.12) with upregulated gene expression encoding GAPDH, while the Rubisco was downregulated (Fig. 7).

Fig. 7
figure 7

Transcriptional regulation on metabolic pathways involving photosynthesis in biofilms based on KEGG pathways. Key enzymes are presented as their names in red colours (upregulated) and in blue colours (downregulated). Cytb6, cytochrome b6/f complex subunit; petH, ferredoxin—NADP+ reductase; LHCA, light-harvesting complex I; LHCB, light-harvesting complex II; ATP synthase, F-type H+ transporting ATPase

Full size image

Discussion

The advantage of manipulative experiments lies in their high controllability, whereas outdoor experiments are more closely natural environments. Compared to in situ experiments in rivers, this approach minimizes interference from human activity and extreme weather, thereby allowing for a more focused study on the mechanisms of shading on the fatty acid composition of biofilms. In our mesocosms, by manipulating light intensity and regulating water flow, we simulated shading and movement of river headwaters or upstream areas with dense canopy cover. We transported cobbles with colonized biofilms from natural rivers to ensure that the biofilms used in our experiments closely reflected the conditions and variation occurring in natural environments. Additionally, the water used in the mesocosms had total dissolved nitrogen (TDN) and total dissolved phosphorus (TDP) similar to those of the river from which the biofilms were collected, thereby ensuring that our findings are both relevant and applicable. To account for local variation in biofilm composition, we addressed it through random collection and composite sampling as previously described. Thus, our study provides a foundation that reveals the mechanisms underlying the impact of shading on basic resources.

Light availability, indirectly influenced by factors such as canopy cover, is a dominant abiotic factor affecting the community structure of benthic algae in biofilms [1, 40]. Lower light conditions can lead to a decrease in the concentration of short-chain PUFA such as LIN and ALA in biofilms, while EPA and ARA tend to increase [41, 42], and higher light intensity leads to excess carbon accumulation in storage lipids, such as SAFA and MUFA [43]. Several studies have debated the effect of light intensity on SAFA% [41, 42]. Even under open canopy conditions, light intensity may vary significantly and may account for the differing responses in SAFA. Our study observed a significant decrease in SAFA% in the shaded group compared to the ambient group, which aligns with the trend reported in a previous study [41]. In addition, SAFA accumulates under high temperature conditions, while PUFA tends to increase to maintain the fluidity of cell membranes under low temperature [44]. Although the water temperature in our mesocosms was higher than in the natural river, there was no significant difference between the shaded and ambient groups (Table 1).

The observed changes in PUFA may be attributed to physiological acclimation, specifically photoacclimation. In photoautotrophic eukaryotic microalgae, LC-PUFA primarily accumulates within complex polar lipids, such as glycolipids and phospholipids, which constitute the structural basis of cellular membranes [45, 46]. Photoautotrophic growth can be limited by light availability [47]. However, photoacclimation to low irradiance involves the development of associated thylakoid membranes, thereby enhancing the absorption of limited photons [41, 48]. Prior research suggests that photoacclimation likely increases PUFA content due to its abundance in thylakoid membranes [49]. In our study, increased PUFA% was consistent with a photoacclimation mechanism, involving the enhanced synthesis or desaturation of thylakoid membranes [41]. This suggests that under shading treatment, algae in biofilms adapt by modifying their membrane composition to optimize light absorption, thereby increasing PUFA accumulation. Studies on marine diatoms have shown a decline in PUFA as light levels rise, further indicating the involvement of photoacclimation [50, 51].

Diverse conditions not only modify the community structure of benthic algae in biofilms but also, to some extent, alter its biochemical composition [42, 52]. Our results support the second hypothesis that the changes in the nutritional quality of biofilms are linked to algal community succession, influenced by shading (Fig. 2, Table 2). Bacillariophyta are typically dominant in near-pristine streams [15], whereas Chlorophyta are more common in open streams exposed to high irradiance [53]. The increased C16:1n7 in biofilms under shading could be explained by the increased abundances of Bacillariophyta (Table 3, Fig. 2) [54, 55]. Shading diminished the competitive advantage of Cyanobacteria within the community (Fig. 2). However, it did not significantly affect the biomarker fatty acid ALA, despite an observed trend that might contribute to its decrease (Table 3). Additionally, although EPA and DHA did not show statistical significance, the increase in LC-PUFA implies an improvement in nutritional quality (Fig. 3a, b). The altered environmental condition provides growth opportunities for other algae, especially those well-adapted to low light, such as diatoms [15].

When algae are exposed to changing environmental conditions and stresses, they frequently adapt by modulating their lipid metabolism pathways, which encompass multiple metabolic routes [56]. While KEGG enrichment analysis showed no significant effects of shading on the overall fatty acid biosynthesis and elongation pathways (Fig. 4), it is essential to highlight the role of specific differentially expressed genes (DEGs) within these pathways. The absence of significant pathway level changes does not negate the importance of key regulatory enzymes, which can still play crucial roles in regulating fatty acid metabolism under shading treatment. During fatty acid biosynthesis, acetyl CoA is a precursor substance for fatty acid biosynthesis, which is converted to malonyl-CoA by acetyl-CoA carboxylase (ACACA, EC: 6.4.1.2, 6.3.4.14). Interestingly, ACACA catalyzing the initial step of de novo FA biosynthesis was upregulated in the shaded groups. However, 3-hydroxyacyl ACP dehydratase (FabZ, EC: 4.2.1.59) was downregulated, suggesting a repressed de novo FA biosynthesis. Although the expression level of long-chain acyl-CoA synthetase (ACSL, EC: 6.2.1.3) in the shaded group was higher, the product of ACSL, known as hexadecanoyl-CoA, could inhibit FA synthesis through allosteric negative feedback [57]. This inhibition may lead to the downregulation of key enzymes involved in de novo fatty acid synthesis. Concurrently, the downregulation of FAB2 and FATA responsible for SAFA (primarily C16:0 and C18:0) synthesis was observed, consistent with the reduction of SAFA (Table 3).

In the cytosol, free FAs can be ligated to CoA via ACSL producing acyl-CoA [58, 59], which could be assembled to the glycerol backbone to form TAG. Since endoplasmic reticulum (ER)-based FA elongation is involved in multiple biological processes providing precursors of membrane lipids and lipid mediators [25], it is important to maintain basal lipid supply to biological processes. The KCS catalyzing the first and rate limiting condensation step of elongating FAs in ER was upregulated significantly under shading (Fig. 5, Table S2), implying that KCS might be the key enzyme controlling PUFA biosynthesis. Under shading, the upregulation of FAD2 drives the conversion of saturated to unsaturated FA (Fig. 5), and such shifting carbon flux towards lipid synthesis may represent a response to stressful conditions [60]. The repression of the FA biosynthesis and enhancing FA elongation imply that the key regulation point of increased PUFA biosynthesis in shading might be not relying on de novo FAs synthesis [23, 61]. Furthermore, GSEA revealed significant activation of pathways involved in fatty acid elongation and arachidonic acid (ARA) metabolism (Fig. 6a, b). This suggests that shading induces the upregulation of PUFA-related synthesis and metabolic pathways, providing further evidence for the increased synthesis of PUFA, particularly ARA.

Photosynthesis also imposes a critical effect on the flux of acetyl CoA, NADPH, and ATP supply toward lipid biosynthesis [62]. However, under shading conditions, the gene expression of petH and ATP synthases were downregulated (Table S4), leading to reduced NADPH and ATP production, which ultimately lowers the energy required for FA synthesis. In contrast, the increased expression of GAPDH indicates the enhancement of the Calvin cycle pathway (Table S4), and the ultimate product G-3-P serves as a carbon source for lipid synthesis. LHC serving as a light receptor efficiently captures and transfers excitation energy to photosystems. The upregulation of LHC-related genes responsive to the shading may enable algae to flexibly make adjustments in photosynthetic efficiency under adverse abiotic stress [63]. Sustaining essential metabolic processes in the shading is imperative to enable rapid photosynthesis upon re-illumination [64].

The repression of glycolysis-related genes, such as pyruvate kinase (PK, EC: 2.7.1.40) (Table S2), suggests that biofilms tend to decrease the consumption of carbohydrates to save energy. This downregulation subsequently hinders the provision of additional substrates for fatty acid biosynthesis under shading conditions [23]. As the end product of glycolysis, pyruvate links glycolysis with TCA cycle and lipid biosynthesis, thus pyruvate is also considered a regulative metabolite for adjusting lipid biosynthesis [65, 66]. Pyruvate can also be converted to acetyl-CoA by a pyruvate dehydrogenase (PDH, EC: 1.2.4.1) bypass pathway, as an alternative pathway supplying acetyl-CoA for enhanced lipid biosynthesis [67]. The downregulation of PDH suggests a reduced conversion of pyruvate to acetyl-CoA and its incorporation into the tricarboxylic acid (TCA) cycle when compared to the ambient groups. Pyruvate might not provide a sufficient amount to replenish the levels of TCA intermediates, which are consistently depleted to meet the increased energy demands for lipid biosynthesis [67].

Conclusion

This study elaborates that nutritional quality indicated by LC-PUFA in stream food webs increases with riparian rehabilitation in a manipulative experiment. Result reveals mechanisms in two aspects contributing to increased nutritional quality: succession in algal community from Cyanobacteria and Chlorophyta to Bacillariophyta and upregulation of key enzyme such as FAD2 encoding genes in PUFA biosynthesis. Photosynthesis indirectly provides acetyl CoA, NADPH, and ATP which serve as precursors and energy for FA biosynthesis reactions. The increased LC-PUFA% at shading conditions suggests that photoacclimation influences the synthesis and desaturation of membranes. Our study provides a comprehensive understanding of stream food webs in terms of basal resource nutritional quality and offered empirical evidence supporting riparian revegetation as a vital strategy for watershed management.