2.1. Preparation of neem extract

Neem kernel extract stock solution (3 g/L) was prepared from a commercial product (Coromandel/Murugappa; Parry’s mBio NeemAzal® Technical Pack No 06/1/3, Lot No B365). This formulation contains azadirachtin A (25–50%) along with other naturally occurring kernel constituents, including additional azadirachtins (14–21%), limonoids (1–10%), and fatty acids (1–2%). These values correspond to the compositional range declared for the commercial product and were not independently quantified in the present study. Therefore, exposure concentrations are reported as nominal concentrations of total neem extract in mg/L or ppm, rather than as analytically measured concentrations of azadirachtin A or individual limonoids. Because of their low solubility in water, dilutions were made in 0.5% dimethyl sulfoxide (DMSO) in artificial seawater. The control group received artificial seawater containing 0.5% DMSO only, corresponding to the solvent concentration used in all neem extract dilutions. The stock was stirred overnight at room temperature before use. Neem concentrations are reported as ppm (mg/L).

2.2. Culture of juvenile and adult Artemia

A. franciscana were maintained in artificial seawater at 35–40 ppt and 28°C under a controlled 14 h light/10 h dark cycle. A. franciscana cysts were hatched under the same salinity, temperature, and photoperiod conditions. Juvenile Artemia were collected 12–24 h post-hatching, corresponding to nauplius instar stages I–II, and placed in 24-well plates with approximately 50 juveniles per well. Artificial seawater was removed, and 2 mL of test solution were added to each well. Test solutions consisted of nominal neem extract dilutions prepared in artificial seawater at 35–40 ppt with 0.5% dimethyl sulfoxide (DMSO), as described above. Control wells contained artificial seawater with 0.5% DMSO only. Nauplii were incubated at neem extract concentrations ranging from 0 to 800 ppm for 24 h at 28°C under a 14 h light/10 h dark cycle and scored after 24 h. Test solutions were not replaced during the 24 h juvenile exposure.

At nauplius instar stages I–II, normal individuals showed active swimming with coordinated undulating movements of the podia (S1 Video; control, 00:04–00:10) and were scored as normal. Individuals showing reduced and/or uncoordinated podial movements and impaired ability to remain in the water column were classified as abnormal (S1 Video; 00:17–00:30). Individuals that remained immobile at the bottom of the well, showed no response to gentle tapping, and exhibited necrotic morphology were classified as dead (S1 Video; 00:30–00:58). The results were expressed as lethal concentration (LC) values for mortality. The effective concentration (EC), representing the percentage of abnormal and deceased animals within the population, was also calculated.

Adult A. franciscana were grown and maintained as a laboratory colony in artificial seawater at 35–40 ppt and 28°C under a controlled 14 h light/10 h dark cycle (S2 Video). Adults used in the exposure assays were sexually mature individuals selected by external reproductive morphology: males were identified by the presence of claspers, and females were selected based on the presence of a visible brood sac. Experimental groups were maintained in six-well culture plates containing 10 mL of test solution per well, with three individuals of the same sex per well. Each concentration (0, 250, and 500 ppm) was tested in six replicate wells (n = 18 adults per sex and treatment). Animals were exposed to nominal neem extract dilutions prepared in artificial seawater at 35–40 ppt with 0.5% DMSO, using the same solvent concentration as in the juvenile assays. Adults were maintained for seven days under the same temperature and photoperiod conditions described above, with daily renewal of the full test solution volume (10 mL per well). Animals were fed daily with SERA® Micron Nature Fry Food (SERA GmbH), supplied as two drops per well from a 1 g/L suspension prepared in artificial seawater. Because no consistent sublethal abnormalities could be reliably scored in adults, mortality was recorded daily throughout the 7-day exposure period.

Overall, the experimental design comprised an acute 24 h juvenile exposure (0–800 ppm) used for toxicity assessment and juvenile RNA-seq in the sublethal range (0–400 ppm), and a 7-day adult exposure that supported three endpoints: survival analysis at 0, 250 and 500 ppm, adult RNA-seq at 0, 250 and 500 ppm, and histology assessments at 0, 100, 250 and 500 ppm.

2.3. Total RNA extraction and construction of cDNA libraries

After the 24-hour exposure to the sublethal range of neem extract (0 ppm control, 100, 200, 300 and 400 ppm), with each treatment performed in triplicate (three wells per concentration; approximately 50 juveniles per well), juveniles were collected from each well using filter paper, preserved in RNAlater (Thermo Fisher Scientific), and stored at −80 ºC. Each well was treated as one biological replicate, yielding three independent RNA samples per juvenile treatment.

Adult male and female Artemia were exposed for seven days to neem extract at 0 ppm (control), 250 ppm, and 500 ppm. For each sex and concentration, three RNA-seq libraries were prepared. Each library consisted of a pool of three surviving adults collected at day 7 from the same sex and treatment group. When necessary, individuals were combined from more than one well within the same sex and treatment group to obtain the number required for library construction.

Total RNA was extracted from pooled brine shrimp using the TRIzol reagent (Invitrogen) protocol. RNA purity was assessed using the A260/A280 ratio with a NanoDrop Lite spectrophotometer (Thermo Fisher Scientific), and RNA integrity was verified by visualizing samples in agarose gels under denaturing conditions. Subsequently, a total of 33 double-stranded cDNA libraries were generated using the TruSeq® RNA Sample Preparation Kit v2 (Illumina®) by Novogene USA, corresponding to three biological replicates per treatment and life stage/sex combination (five juvenile treatments × three replicates, and three adult treatments × two sexes × three replicates). All libraries were sequenced on the NovaSeq 6000 platform (Illumina®) with 2 × 150 bp paired-end reads. The list of sequenced cDNA libraries is detailed in S1 Table. The original sequencing files have been uploaded to NCBI’s Sequence Read Archive (SRA) under the BioProject ID PRJNA1402122.

2.4. Sequencing of cDNA libraries and bioinformatics analysis

Transcriptomic analyses were conducted as an exploratory transcript-level assessment of treatment-associated abundance patterns using a de novo assembled reference transcriptome generated from the RNA-seq libraries included in this study. De novo transcriptome assembly can provide a useful reference for differential abundance analyses in emerging or non-conventional model systems, particularly when the objective is to generate a transcript-level resource from the biological material under study [24]. After adapter removal and quality trimming, clean reads were assembled de novo into a unified transcriptome using CLC Genomics Workbench software v11.0.1 (CLC Bio). The assembly parameters included a mismatch cost of 2, an insert cost of 3, a minimum contig length of 400 bp, a similarity threshold of 0.9, and a trimming quality score of 0.05. This assembly yielded 88,367 contigs that were subsequently annotated to the NCBI non-redundant and UniProtKB/SwissProt Protein databases using the BLASTx tool. The de novo assembled transcriptome and the corresponding BLAST-based annotation table are publicly available in BioStudies under accession S-BSST3203. Because this analysis was based on a de novo transcriptome assembly, different contigs may represent fragmented transcripts, alternative isoforms, paralogous sequences, redundant assemblies of the same gene, or chimeric sequences [25]. Accordingly, expression results are interpreted at the transcript/contig level rather than as definitive gene-level estimates.

The annotated transcriptome served as a reference for differential expression analysis by RNA-seq. Clean sequences obtained from each library were aligned to the reference transcriptome using CLC Genomics Workbench software. Expression values were calculated as transcripts per million mapped reads (TPM). The complete juvenile and adult transcript-abundance datasets are publicly available in BioStudies under accession S-BSST3203. Differentially expressed contigs meeting the criteria of |log2FC| ≥ 4 and a false discovery rate (FDR) p-value < 0.05 were identified and visualized in hierarchical clustering heat maps. Hierarchical clustering employed Euclidean distance and full linkage as parameters for analysis.

Gene Ontology (GO) enrichment analysis was performed after differential expression analysis using the Rank-based Gene Ontology Analysis with Adaptive Clustering (RBGOA) tool [26]. GO categories related to molecular function, biological process, and cellular component were evaluated using the differential-expression results, including Log₂ Fold Change (log2FC) and p-value information [27]. Only annotated sequences with an E-value < 0.01 against UniProt entries were retained for GO-based interpretation.

2.5. In silico identification of genes associated with chitin synthesis

To investigate chitin synthesis in A. franciscana, a gene panel was curated from the assembled transcriptome, selecting contigs associated with this process through bioinformatic analysis and literature search on the genes and pathways associated with chitin metabolism in various crustacean species [28]. Homology searches against the NR protein database (E-value < 0.005) identified sequences related to chitin synthesis. These gene panels were used as reference datasets and reads from larvae and adults across treatment replicates were mapped against them using CLC Genomics Workbench, applying the same RNA-seq settings and hierarchical clustering parameters as described previously. Expression levels for the selected genes were quantified as transcripts per million (TPM) derived from the mapped libraries, and these values were used for subsequent analyses.

2.6. Histological analysis of adult A. franciscana

A total of 112 adult A. franciscana were selected for histological evaluation (14 males and 14 females per concentration), and these individuals were reserved exclusively for histological processing. Animals were fixed in Bouin's solution for 48 h and subsequently rinsed under running tap water. Standard histological procedures were followed, including dehydration through a graded ethanol series, clearing in butanol, and embedding in Paraplast Plus® (Merck). Embedded tissues were sectioned at 5 μm thickness using a Leica RM2255 microtome and mounted on glass slides. For visualization, tissue sections were dewaxed and rehydrated via a xylene-butanol sequence and processed for trichrome staining: Hematoxylin (Merck) for nuclei, Erythrosin-orange G (Merck) for cytoplasmic differentiation, and Aniline blue (Merck) for connective tissues and secretion granules. Slides were dehydrated in ethanol, cleared in xylol (Merck), and mounted in Entellan™ (Merck). A Leitz-Leica DMRBE microscope equipped with a Leica DFC290 digital camera was used to document histological features.

The gonadal areas of male and female individuals were identified and photographed, with images taken from at least three consecutive sections per individual when available. Gonadal area measurements were performed from digital micrographs using ImageJ software (version 1.54h; National Institutes of Health) after calibration with the microscope scale bar. Areas were quantified from consecutive sections per individual and averaged for downstream analysis. Representative histological images, including EBS-associated alterations in treated females, are presented in the main figures, and additional quantitative data on gonadal area measurements are provided in the supplementary material.

2.7. Statistical analysis

Statistical analyses were conducted to evaluate the effects of neem extract exposure on A. franciscana at phenotypic, transcriptomic, and histological levels. Phenotypic data, including mortality and developmental abnormalities in juveniles, were analyzed using one-way analysis of variance (ANOVA), followed by Tukey’s multiple comparison test to determine significant differences between groups (P < 0.05). For adult survival analysis, a Kaplan-Meier survival plot was generated using GraphPad Prism software v10.4.1 (GraphPad Software), with statistical differences evaluated using the log-rank (Mantel-Cox) test in the same software.

For transcriptomic analysis, Principal Component Analysis (PCA) was performed on transcripts per million (TPM) values derived from paired-end cDNA libraries, projecting samples onto a two-dimensional space defined by the first and second principal components of the covariance matrix. Differential gene expression analysis was conducted using a T-test with false discovery rate (FDR) correction for multiple comparisons. Independent filtering was applied using a cutoff of |log2FC| ≥ 4 to minimize false positives, and a final subset of transcripts was retained for further analysis, with statistical significance set at P < 0.05. All RNA-seq statistical analyses were performed using CLC Genomics Workbench software.

For the expression analysis of chitin-related genes, TPM values were used as input for statistical comparisons, and gene expression differences in juveniles were assessed using one-way ANOVA, followed by Tukey’s multiple comparison test (P < 0.05). In adults, chitin-related gene expression was analyzed using two-way ANOVA, followed by Tukey’s multiple comparison test to evaluate differences between treatment groups (P < 0.05).

For histometric data, normality was assessed using the Shapiro–Wilk test. In cases where data did not meet normality assumptions, Box-Cox transformation was applied [29]. Statistical comparisons between experimental conditions were performed using one-way ANOVA, followed by Tukey’s multiple comparison test (P < 0.05).

All statistical analyses for phenotypic data, chitin-related gene expression, and histometric data were conducted using GraphPad Prism software (10.4.1).

2.8 Ethics statement

This study involved Artemia franciscana, a non-cephalopod invertebrate, and did not involve human participants, vertebrate animals, cephalopods, or field research. Therefore, institutional ethics approval was not required.

2.9. Artificial intelligence tools and technologies

During manuscript preparation, the authors used ChatGPT (OpenAI) to assist with language editing and to improve the clarity and consistency of the manuscript text. All AI-assisted output was reviewed and edited by the authors for accuracy and consistency with the original results and interpretations. No primary data, analyses, results, figures, tables, or Supporting Information files were generated by the AI tool. The authors take full responsibility for the content of the manuscript.

3.1. Neem extract induces dose-dependent mortality and abnormal phenotypes in juvenile A. franciscana

At 24 h, all juveniles in the control group (0 ppm) exhibited a normal phenotype (Fig 1A). At concentrations below 100 ppm, most juveniles remained normal. Between 100 and 300 ppm, the proportion of abnormal juveniles increased significantly (P < 0.001), accompanied by a marked reduction in the proportion of normal individuals. At 200 ppm, only ~17% of juveniles remained normal, whereas abnormal and dead individuals accounted for the majority of the population. The proportion of abnormal juveniles peaked at 300 ppm (~64%) and then declined at higher concentrations as mortality became the predominant response. Mortality increased markedly from 400 ppm onward, exceeded 85% at 600 ppm, and reached approximately 95% at 800 ppm. Based on these concentration-response data, the 24 h LC₅₀ for juvenile mortality was 292.48 ppm (95% CI: 257.75–331.89), and the EC₅₀ for the combined endpoint “abnormal + dead” was 146.36 ppm. The replicate-level juvenile phenotype counts and percentages used to generate Fig 1A are provided in S1 Dataset.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Dose-dependent effects of neem extract on survival and developmental phenotype of Artemia franciscana.

(A) Juveniles exposed for 24 h. The proportion of individuals displaying a normal phenotype (green) decreased with increasing neem concentration, while abnormal phenotypes (orange) increased, peaking at 300 ppm, and declined at higher concentrations as mortality (red) increased. Data are shown as mean ± SD. Statistical differences were evaluated using one-way ANOVA followed by Tukey’s multiple comparison tests. (B, C) Survival of adult Artemia. Kaplan–Meier survival curves are shown for (B) males and (C) females exposed to 0, 250, and 500 ppm neem. Statistical differences were evaluated using the log-rank (Mantel–Cox) test.

https://doi.org/10.1371/journal.pone.0348463.g001

3.2. Neem extract exposure reduces adult survival in A. franciscana

Preliminary observations indicated that adult A. franciscana were more resistant to neem extract than juveniles, which exhibited phenotypic effects within 24 hours of exposure. Therefore, the potential effects of neem extract on adults were assessed over a period of 7 days. Males and females were analyzed separately, with males identified by the presence of claspers, and only gravid females included in the study (S2 Video). At the lower concentration tested (250 ppm), some mortality was observed by the end of the 7-day exposure, but the difference from controls was not statistically significant. After seven days of exposure to 500 ppm, both males and females showed pronounced mortality (>80% in each sex; Fig 1B for males, Fig 1C for females), whereas control groups exhibited low mortality (approximately 11% by day 7, with deaths beginning on day 5). At both 250 and 500 ppm, male survival declined earlier than female survival in the plotted trajectories. During the exposure period, females in all groups continued to release free-swimming nauplii (ovoviviparous reproduction). The daily adult survival counts used to generate Fig 1B and Fig 1C are provided in S1 Dataset.

3.3. RNA-seq reveals widespread gene expression changes in neem-treated A. franciscana

To explore transcript-level responses associated with neem exposure, we performed a de novo high-throughput transcriptome assembly to serve as a reference for differential expression analysis. After quality filtering, the sequencing yielded 1,604,747,440 clean reads (Table 1), which assembled into 88,367 contigs with an N50 of 1,120 bp. About 70% of contigs (61,999 sequences) were successfully annotated against protein databases, which served as the reference for differential expression analysis.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Summary of RNA-seq output and de novo transcriptome assembly statistics for Artemia franciscana. Values are based on 33 cDNA libraries from juveniles and adults and include total raw and clean reads, mean read length, number of contigs, total assembled bases, mean contig length, N50 and GC content.

https://doi.org/10.1371/journal.pone.0348463.t001

The RNA-seq design included concentration-specific treatment groups. Juvenile libraries represented 0, 100, 200, 300, and 400 ppm neem extract, whereas adult libraries represented 0, 250, and 500 ppm neem extract in males and females. Across all samples, the exploratory screening identified a broad set of candidate treatment-responsive transcripts. Principal component analysis (PCA) of the global gene expression profiles (Fig 2) revealed a clear separation between juveniles and adults along the first principal component (Fig 2A). When analyzed within each life stage, control samples separated from neem-exposed samples. In juveniles, exposed groups separated from controls along PC1 (Fig 2B), with the 100 and 200 ppm groups clustering closer to each other and the 300 and 400 ppm groups showing partially overlapping but distinct profiles. In adults, transcript abundance profiles also diverged between control and exposed individuals (Fig 2C), while PC2 separated males from females, consistent with sex-specific transcriptomic differences in Artemia. Notably, the number of neem-responsive differentially-expressed contigs was much higher in adults (19,957) than in juveniles (933).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Principal component analysis (PCA) of global gene expression profiles in A. franciscana exposed to neem extracts.

(A) PCA of all RNA-seq libraries (n = 33) showing separation between juveniles and adults along the first principal component (PC1). Each point represents one library; colours and/or symbols indicate life stage, sex and treatment (see panel legend). (B) PCA of juvenile libraries only (0, 100, 200, 300 and 400 ppm of neem extract), showing separation between control and neem-treated groups along PC1. (C) PCA of adult libraries only (0, 250 and 500 ppm of neem extract), illustrating differences between control and treated samples and separation between males and females.

https://doi.org/10.1371/journal.pone.0348463.g002

3.4. Transcript abundance patterns in juvenile A. franciscana exposed to neem are consistent with metabolic and cuticle-associated candidate responses

Hierarchical clustering of differentially expressed contigs revealed distinct transcriptional differences between untreated and neem-treated juvenile A. franciscana, with control samples forming a separate cluster from all exposed groups (Fig 3A). Among neem-treated juveniles, the 100 and 200 ppm groups clustered together, whereas the 300 and 400 ppm groups showed partially overlapping but distinct profiles.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Transcriptomic responses to neem extract exposure in juvenile A. franciscana.

(A) Heat map of hierarchical clustering of differentially expressed transcripts from juvenile A. franciscana exposed for 24 h to 0 (control) or 100, 200, 300 and 400 ppm neem (n = 3 libraries per treatment; 15 libraries in total). Columns represent individual libraries (controls in green; neem-treated groups in progressively darker warm colours from 100 to 400 ppm). The colour scale indicates relative transcript abundance, from lower (blue) to higher (yellow) expression. (B) Enriched GO categories for molecular function in treated (left) and untreated (right) juveniles. Bar length indicates the number of differentially-expressed transcripts assigned to each GO term. “Total gene fraction” denotes the proportion of all annotated transcripts in the transcriptome associated with that term. Only terms with P < 0.01 are shown. The colour gradient reflects false discovery rate (FDR), with blue indicating lower FDR (higher confidence) and red indicating higher FDR.

https://doi.org/10.1371/journal.pone.0348463.g003

Differential expression analysis (Table 2) identified a set of candidate contigs with pronounced changes in expression following neem exposure. Strongly induced genes included pseudouridine-5′-phosphatase (log₂FC = 3.04), an enzyme involved in RNA modification, and GTP cyclohydrolase 1 (log₂FC = 2.99), which participates in folate biosynthesis. Other upregulated transcripts included cathepsin S (log₂FC = 2.97), a lysosomal protease associated with protein catabolism, and mitochondrial uncoupling protein 3 (log₂FC = 2.87), a protein linked to regulation of mitochondrial membrane potential.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Differentially expressed transcripts in juvenile Artemia franciscana exposed for 24 h to neem extract, compared with untreated controls. The table lists the top 10 upregulated and top 10 downregulated transcripts, ranked by |log₂ fold change|. For each contig, UniProt accession, lowest BLASTx E-value, FDR-adjusted p-value and log₂ fold change are shown together with the corresponding protein annotation.

https://doi.org/10.1371/journal.pone.0348463.t002

Several transcripts encoding enzymes related to metabolism and detoxification showed marked reductions in expression levels. These included esterase FE4 (log₂FC = −4.24) and carboxylesterase 4A (log₂FC = −3.80), both esterases commonly associated with xenobiotic processing, as well as an acetyl-CoA synthetase 2-like enzyme (log₂FC = −3.56), involved in acyl-CoA formation. Transcripts annotated as lactoperoxidase (log₂FC = −3.23), a heme-containing oxidoreductase, and a pancreatic lipase-related protein (log₂FC = −3.22), implicated in lipid hydrolysis, were also significantly downregulated in neem-treated juveniles.

Gene Ontology (GO) enrichment analysis in juveniles (Fig 3B) showed significant over-representation of metabolic, immune, and structural functional categories. For biological processes, carbohydrate and polysaccharide catabolic processes were among the most strongly enriched terms. Immune-related categories were also enriched, including terms associated with innate and stress-related immune responses. For molecular function, the term “structural constituent of cuticle” showed the highest enrichment, reflecting consistent changes in genes encoding cuticle-associated proteins.

3.5. Targeted analysis of chitin-related transcripts in juvenile A. franciscana

In addition to the global transcript-level analysis, we inspected the abundance of a predefined panel of eight candidate contigs annotated as related to chitin precursor metabolism and chitin-associated processes in juvenile A. franciscana. This analysis was based on TPM-derived abundance estimates from the same RNA-seq dataset and is therefore interpreted as an exploratory candidate-transcript analysis. Seven of these genes were significantly downregulated in neem-treated juveniles compared with controls. The strongest reductions were observed for hexose phosphate aminotransferase 1 and chitin synthase chs2 (Fig 4A,4B), which encode enzymes involved in precursor production and chitin polymerization, respectively. Overall, neem-treated juveniles showed lower transcript abundance for several chitin-pathway components during early developmental stages.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Expression of chitin-related transcripts in juvenile A. franciscana exposed to neem extract.

Transcripts per million (TPM) for selected contigs involved in chitin precursor production and chitin polymerization in juveniles exposed for 24 h to 0, 100, 200, 300, and 400 ppm neem extract. Panels show (A) hexosephosphate aminotransferase 1, (B) chitin synthase chs-2, (C) phosphoglucomutase, (D) hexokinase type 2, (E) phosphoglucomutase-3, (F) alpha-trehalose glucohydrolase, (G) glycogen phosphorylase, and (H) phosphohexose isomerase. Bars show mean ± SEM (n = 3 libraries per treatment). Different lowercase letters indicate significant differences among treatments (one-way ANOVA followed by Tukey’s multiple comparison test).

https://doi.org/10.1371/journal.pone.0348463.g004

3.6. Adult transcript abundance patterns indicate stress-, metabolism-, and structural candidate responses to neem exposure

Exposure to neem extract was associated with transcript abundance differences in adult A. franciscana. Hierarchical clustering of differentially-expressed transcripts showed that untreated samples generally formed a distinct cluster from treated groups (Fig 5A). One sample from the 500 ppm treatment (AFATT2_1_1) grouped with controls, indicating within-group variation. Despite this, most treated samples exhibited transcriptional shifts compared to controls. Within the treated groups, adults exposed to 500 ppm displayed a transcriptional profile that differed from both the 250 ppm group and untreated controls.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. Transcriptomic responses to neem extract exposure in adult A. franciscana.

(A) Heat map of hierarchical clustering of differentially-expressed transcripts from adult A. franciscana exposed for 7 days to 0 (control), 250 and 500 ppm neem (n = 3 libraries per sex and treatment). Colours indicate relative transcript abundance (low to high). (B–D) Enriched Gene Ontology (GO) categories for molecular function (B), biological process (C) and cellular component (D) in treated (left) and untreated (right) adults. Bar length indicates the number of DEGs assigned to each GO term. “Total gene fraction” denotes the proportion of all annotated contigs in the transcriptome associated with that term. Only terms with P < 0.01 are shown. The colour gradient reflects false discovery rate (FDR), with blue indicating lower FDR (higher confidence) and red indicating higher FDR.

https://doi.org/10.1371/journal.pone.0348463.g005

Differential expression analysis (Table 3) identified several transcripts with large changes in expression in neem-exposed adults. Strongly upregulated transcripts included elongation factor 1-alpha (log₂FC = 14.09), a translation factor involved in protein synthesis, and tubulin alpha chain (log₂FC = 13.54), a major component of the cytoskeleton. Heat shock protein 70 protein 4 (HSPA4) (log₂FC = 12.07), a molecular chaperone of the HSP70 family, was also highly induced, as was a major cysteine proteinase (log₂FC = 13.18), a protease associated with protein turnover.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Differentially-expressed transcripts in adult Artemia franciscana exposed for 7 days to neem extract, compared with untreated controls. The table lists the top 10 upregulated and top 10 downregulated transcripts, ranked by |log₂ fold change|. For each contig, UniProt accession, lowest BLASTx E-value, FDR-adjusted p-value and log₂ fold change are shown together with the corresponding protein annotation.

https://doi.org/10.1371/journal.pone.0348463.t003

Several transcripts associated with energy metabolism and other homeostatic functions showed reduced expression levels in adults exposed to neem. Glycerol-3-phosphate dehydrogenase, an enzyme involved in glycolysis and lipid metabolism, was strongly downregulated (log₂FC = −4.86), as was carbohydrate sulfotransferase 11 (log₂FC = −6.65), which participates in glycosaminoglycan biosynthesis.

Gene Ontology (GO) enrichment analysis in adults (Fig 5B–5D) showed significant over-representation of categories related to “regulation of gene expression” and “transcription factor activity” in treated samples. In contrast, untreated adults exhibited higher representation of GO terms such as “ribosomal large subunit assembly” and “rRNA processing”, as well as lipid metabolic processes including lipase and aldehyde dehydrogenase activities. Treated adults also showed enrichment of structural molecule activity associated with the extracellular region and cuticle.

Collectively, these transcript abundance and GO patterns indicate candidate stress-, metabolism- and cuticle-associated responses to neem exposure in adults.

3.7. Exploration of candidate chitin-related transcript expression in adult A. franciscana exposed to neem

To determine whether chitin-associated pathways were also affected in adults, we analysed the same panel of eight genes related to chitin biosynthesis and degradation used for juveniles in neem-treated and untreated adult A. franciscana (Fig 6). Three of these transcripts showed significant expression differences between treatments. In males, hexokinase type 2 and phosphohexose isomerase, both associated with carbohydrate metabolism and chitin precursor synthesis, were significantly downregulated in neem-exposed individuals compared with controls (Fig 6A,6B). In females, glycogen phosphorylase, an enzyme contributing to mobilization of stored carbohydrates toward biosynthetic pathways, showed reduced expression in treated samples (Fig 6C).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. Expression of chitin-related transcripts in adult A. franciscana exposed to neem extract.

Transcripts per million (TPM) for selected contigs involved in chitin precursor metabolism in adult males and females exposed for 7 days to 0, 250, and 500 ppm neem extract. Panels show (A) hexokinase type 2, (B) phosphohexose isomerase, (C) glycogen phosphorylase, (D) alpha-trehalose glucohydrolase, (E) chitin synthase chs-2, (F) phosphoglucomutase-3, (G) glucose phosphomutase, and (H) hexosephosphate aminotransferase 1. Bars show mean ± SEM (n = 3 libraries per sex and treatment). Brackets indicate the pairwise comparisons tested, and asterisks indicate statistically significant differences between groups (*P < 0.05, **P < 0.01; two-way ANOVA followed by Tukey’s multiple comparison test).

https://doi.org/10.1371/journal.pone.0348463.g006

3.8. Histological analysis of adult Artemia

Histological analysis was used to evaluate reproductive tissue morphology in adult A. franciscana exposed to neem extract. In untreated control females (0 ppm; Fig 7A–7C), the embryo-bearing brood sac region showed a defined cuticular boundary enclosing the embryos; the moon-shaped embryos within the brood sac corresponded to desiccating A. franciscana cysts (arrows) and exhibited normal morphology. In females exposed to 100 and 250 ppm neem extract (Fig 7D–7I), representative sections showed altered organization of this region, including irregular contour of the cuticular boundary and less sharply delimited embryo-associated structures relative to controls. In females exposed to 500 ppm (Fig 7J–7L), the embryo-bearing brood sac region showed more evident loss of structural organization, including discontinuous or poorly delimited areas of the cuticular boundary and irregular arrangement of embryo-associated structures. Cuticle thickness was not quantitatively assessed in this analysis.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 7. Histological sections of the embryo-bearing brood sac region in the adult female reproductive tract of Artemia franciscana.

Arrowheads indicate the cuticular boundary surrounding embryo-associated structures in the brood sac region. Arrows indicate desiccating Artemia cysts within the brood sac. (A–C) Untreated control female. (D–F) Female exposed to 100 ppm neem extract. (G–I) Female exposed to 250 ppm neem extract. (J–L) Female exposed to 500 ppm neem extract. Representative treated sections show altered organization of the embryo-bearing brood sac region, including irregular or less sharply delimited areas of the cuticular boundary and embryo-associated structures. Moon-shaped structures within the brood sac correspond to Artemia cysts in the process of desiccation. (A, D, G, J): Bar = 200 µm. (B, E, H, K): Bar = 100 µm. (C, F, I, L): Bar = 50 µm.

https://doi.org/10.1371/journal.pone.0348463.g007

Female ovarian area was quantified as a histometric endpoint, whereas direct reproductive output, including released cysts or nauplii, was not measured. Beyond the brood sac region, morphometric analysis showed that females from higher neem treatments had a significantly reduced ovarian area (P = 0.0006) compared to controls (S1 Fig). In contrast, males did not show significant changes in testis area. The gonadal area measurements used to generate S1 Fig are provided in S1 Dataset.