Omics-Based Investigation on Mechanisms Controlling Cellular Internalization of Keratin Monomers during Biodegradation by Stenotrophomonas maltophilia DHHJ Open Access

The widely existing natural outer protective tissues in the animal kingdom, such as feathers, wool, and claws, constitute a large category of natural materials, with their main component being keratin. Keratin is a fibrous and non-nutritive type of hard protein that is difficult for organisms to utilize directly. Therefore, they are disposed of as waste in practice, causing a serious environmental problem. However, the degradation products of keratin – small peptides and amino acids – can be valuable for conversion into fertilizers, feed, chemical raw materials, and more [1]. Keratin is difficult to degrade in nature due to its high cross-linking disulfide bonds, hydrogen bonds, and hydrophobic interactions, which result in high stability. Some techniques such as mechanical [2], chemical [3], and biological degradation methods [4] have been used to degrade keratin. Although mechanical and chemical methods for extracting soluble keratin are possible, these processes can disrupt the internal structure of amino acids, reducing the quality of the degradation products. Additionally, these processes tend to cause high energy consumption and environmental pollution.

Biological process (BP) with microbial enzymes to degrade keratin has the characters such as mild reaction conditions, minimal environmental pollution, and desirable yield of amino acids. Keratin degrading through bioprocessing can produce high-quality and stable products. Currently, multiple microorganisms capable of degrading keratin have been discovered [5]. Understanding the mechanisms of microbial keratin degradation allows for improvement in degradation efficiency through synthetic biology modifications of microorganisms. Keratin is rich in disulfide bonds, and to render it soluble, it is necessary to break the internal disulfide bonds, hydrogen bonds, and hydrophobic interactions. There are various ways to break disulfide bonds in keratin, including sulfur reduction [6], physical pressure treatment, and enzymatic digestion [7], which are among the primary ones. It is important to note that microbial keratin degradation involves multiple factors and is a multi-step process, making the elucidation of the keratin degradation mechanism complex.

Omics approaches allow for a global and systematic analysis of cellular metabolism and processes under specific conditions, which offer valuable tools for deciphering complex BPs, particularly guiding research on the microbial degradation of complex organic compounds. Researchers have developed different omics approaches based on their specific research targets [8]: genomics focusing on DNA variation, transcriptomics aiming at RNA expression, and proteomics targeting protein alteration.

Due to their deep sequencing capabilities, RNA-seq, as a form of transcriptomics, can provide a more comprehensive view of global gene expression in specific tissues and at specific moments in the life of a biological organism [9]. RNA-seq technology predominantly focuses on coding regions. With a reference genome in place, transcriptome sequencing can rapidly yield gene expression profiles and analyze expression levels as well as transcript structures. This greatly enhances our understanding of bacterial gene regulation [10], subsequently aiding in the comprehension of complex metabolic processes. Co-immunoprecipitation (Co-IP) is a classic method for studying protein-protein interactions, primarily used to identify novel partners that interact with a specific protein. When combined with nano-LC-MS/MS, this technique can accurately identify multiple protein components from protein gel bands. Bong-Kwan Phee and others [11] used Co-IP combined with mass spectrometry to determine interacting proteins related to photosensitive pigments in Arabidopsis. Benjamin Free and colleagues [12] applied this method to identify proteins interacting with neurotransmitter receptors.

We previously isolated a bacterium, Stenotrophomonas maltophilia, capable of efficiently degrading feather keratin [13]. In the earlier studies, it was believed that this bacterium initially degrades feathers into water-soluble keratin monomers extracellularly, followed by intracellular monomer decomposition, subsequently initiating the degradation system [14]. Based on these studies, here we inoculated S. maltophilia cells with keratin monomers and analyzed the bacterium’s transcriptome using RNA-seq sequencing. Additionally, by using keratin monomers as bait, we employed Co-IP with LC-MS/MS technology to identify proteins interacting with keratin monomers. The two high-throughput screening techniques were employed to investigate the mechanism of keratin degradation. To our knowledge, this is the first instance of utilizing omics to study the mechanism of keratin biological degradation.

S. maltophilia DHHJ was cultured in beef extract-peptone medium (BPM, 5 g/L beef extract, 10 g/L peptone, 10 g/L NaCl, pH 7.0–7.2) with shaking at 35°C for 14 h as the feed liquid. For the RNA-seq experiment, 1% S. maltophilia DHHJ feed liquid was added in M9 keratin medium (17.1 g/L Na₂HPO₄·12 H₂O, 0.3 g/L KH₂PO₄-0.5 g, 0.5 g/L NaCl, 0.4 g/L glucose, 0.024 g/L MgSO₄, 0.0011 g/L CaCl₂, and 43.2 mg/L keratin, pH 7.5) and cultured with shaking at 35°C for 48 h, named as K, while S. maltophilia DHHJ were cultured in BPM medium with shaking at 35°C for 14 h, named as S.

Total RNA was extracted using RNeasy Mini Kit (Cat# 74106, Qiagen) following the manufacturer’s instructions and checked for RIN number to inspect RNA integrity by an Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, USA). Qualified total RNA was further purified by RNAClean XP Kit (Cat A63987, Beckman Coulter, Inc., Kraemer Boulevard Brea, CA, USA) and RNase-Free DNase Set (Cat#79254, QIAGEN GmBH, Germany).

This protocol describes bacterial RNA sequencing from total RNA. The protocol for library preparation includes the following steps: first, the ribosomal RNA (rRNA) was depleted using the Ribo-Zero^™ rRNA Removal Kit (bacteria), which is based on selective binding of biotinylated probes to rRNAs using a hybridization/bead capture procedure. The remaining mRNAs were then fragmented and reverse transcribed using random primers with an Illumina adapter overhang, and an adapter was ligated to the 3’ ends of the cDNAs. The cDNAs were PCR amplified using indexed primers to allow multiplexing. Finally, the DNA libraries were quantified and pooled, and 300 base pair fragments were selected for sequencing. Cells of K and S cDNA sequencing were operated with Illumina HiSeq 2000/2500.

The cDNA of feather keratin gene (Gene ID: 769269, F-KER, 297 base pair, 98 aa) was synthesized and cloned in pET-32(+) vectors, fused with TrxA, 6×His, and S-tag. The recombinant keratin was digested with thrombin and purified with Ni column affinity chromatography attached with 6×His TrxA. S-tag keratin containing permeate liquid was collected.

In Co-IP experiment, 5 mL of overnight cultured S. maltophilia DHHJ was harvested, and the cell pellet was washed twice with PBS. The cells were then resuspended in 4 mL of IB (10 mmol/L Tris-HCl) buffer, to which 1 mL of S-tag keratin was added. The incubation took place in a 35°C water bath for 1 h. Afterward, the incubated cells were collected and washed twice with PBS. They were then lysed with 200 μL lysis buffer containing 2 μL protease inhibitor and left on ice for 15 min, and the supernatant was collected as bacteria lysate, which was incubated with 2 μL of mouse anti-S-tag monoclonal antibody overnight at 4°C.

To equilibrate the collection column, 100 μL of lysis buffer was added, followed by a spin at 1,000 g for 1 min. A total of 200 μL of the antibody-incubated bacterial cell lysate was loaded to the well-balanced collection column. After a centrifuge at 1,000 g for 1 min, the antibodies in the lysate were retained on the collection column. Then, 100 μL of washing buffer was applied to the collection column, followed by a spin at 1,000 g for 1 min to eliminate the non-specifically bound proteins. To elute the proteins interacting with keratin, 30 μL of elution buffer was added to the collection column, which was then placed in a collection tube containing 3 μL of neutralization buffer (50 mmol/L Tris-HCl). The mixture was centrifuged at 1,000 g for 1 min to collect the eluate. Silver-stained SDS-PAGE and Western blotting were used to verify the collected samples.

Co-immunoprecipitated samples were freeze-dried, in which 40 μL trypsin buffer was applied to enzymatically digest all proteins by incubating at 37°C for 16 h. The enzymatic hydrolysis samples were trapped on Zorbax 300SB-C18 column (Agilent Technologies, Wilmington, DE, USA) and separated by a Zorbax 300SB-C18 capillary column. The separation was conducted with gradient elution using solvent A (0.1% formic acid) and solvent B (0.1% formic acid and 84% acetonitrile), which was run at 4% to 50% solvent B over 50 min, then from 50% to 100% for 5 min, and finally 100% solvent B for 5 min. The resulting peptides after HPLC separation were determined by using Q Exactive mass spectrometer (Thermo Fisher) with a process time of 60 min. The data from mass spectrometry results were analyzed by using Mascot 2.2 to search the S. maltophilia DHHJ protein library and identify proteins that interact with S-keratin.

The differentially expressed genes from the RNA-seq and Co-IP analysis were selected for validation using real-time quantitative PCR (qRT-PCR), which included Ton B-dependent receptors (JJL50_15620 and JJL50_17955), ABC transporter ATP-binding protein (JJL50_03260), and ABC transporter substrate-binding protein (JJL50_20035) (the primer sequences are listed in online suppl. Table S1; for all online suppl. material, see https://doi.org/10.1159/000540072). S. maltophilia DHHJ was cultured (in duplicates) in BPM medium for 14 h and M9 keratin medium for 48 h at 35°C. The cells were pelleted and subjected to lysozyme treatment (200 mg/mL) for 2 h at 37°C, and thereafter, total RNA was extracted by using TRIzol reagent. Isolated RNA (10 μg) from each sample was subjected to DNase I treatment to remove genomic DNA contamination. The purified RNA (1 μg) was used for cDNA synthesis (cDNA synthesis kit; Bio-Rad, USA) as described by manufacturer. 16S rRNA coding gene was used as the housekeeping control. The log2 fold change of four selected genes from both transcriptome and qRT-PCR data was compared by running regression analysis.

We analyzed the differential gene expression in S. maltophilia cells cultured with different nitrogen sources (beef extract and feather keratin) and found that out of over 4,000 genes in both samples, a total of 1,763 genes exhibited differential expression (different expression genes). These differentially expressed genes were categorized into three levels: upregulated, downregulated, and no significant difference (as shown in Fig. 1a). The red diagonal bars in the figure represent the total number of genes upregulated relative to sample S, the green diagonal bars represent the total number of downregulated genes, and the gray horizontal bars represent the total number of genes with no significant change in expression. Among these genes, in the S. maltophilia DHHJ strain induced by keratin, 1,453 genes were upregulated, while 310 genes were downregulated. The number of upregulated genes far exceeded the number of downregulated genes. Furthermore, the fold change in expression for upregulated genes was higher than that for downregulated genes. When the fold change was 6, there were still 33 upregulated genes that met this criterion, whereas there were no genes that met this criterion for downregulated genes with a fold change of 6. Differential gene analysis between samples was conducted using edgeR software package, and differential expression fold changes were calculated based on FPKM values. Significant selection was performed with conditions of fold change ≥2 and q value ≤0.05. As shown in Figures 1b and c, the volcano plot of differential genes and the scatter plot of expression correlation, a total of 1,108 genes met these conditions, all of which exhibited an upregulation trend. From Figure 1b, it can be observed that there are many points deviating from the diagonal line (whether upregulated or downregulated genes), indicating low correlation and poor repeatability in the expression levels between the two samples. This suggests that many genes are involved in the process of keratin degradation in S. maltophilia. In the volcano plot (Fig. 1c), it is evident that the number of upregulated genes is significantly higher than the number of downregulated genes, and the upregulated genes also show higher variability. The scatter plot of differentially expressed genes between the two samples can be used to compare the consistency between samples.

Gene Ontology (GO) analysis was conducted on the differentially expressed genes in S. maltophilia induced by keratin monomers and beef extract. A total of 1,389 genes were annotated into three primary functional categories, which could be further divided into 41 secondary GO classification units. Differential gene expression related to the strain’s keratinase activity involved multiple genes (Fig. 2). These differential genes expressed in BPs are most associated with cellular processes, metabolic processes, single-organism processes, catalytic activity, binding, and transporter activity functions. Additionally, in genes related to cellular components and molecular functions, there was a significant increase in the number of differentially expressed genes related to cell and cell membrane components, binding, transport, and catalysis. Based on the results of GO enrichment analysis, the selected differentially expressed genes were mapped to various entries (terms) in the GO database. The number of genes for each entry was calculated, and then, a hypergeometric test was applied to select GO entries significantly enriched in differentially expressed genes compared to the entire genome background. A threshold of p value ≤0.05 was used to select significantly enriched GO entries (Fig. 3a). The x-axis represents the degree of enrichment (Rich factor), with a larger Rich factor indicating higher enrichment. The y-axis represents the enriched GO terms. Further details of gene expression within each term are shown in Figure 3b. Analysis and comparison revealed that, except for the tricarboxylic acid cycle, more genes were upregulated than downregulated in other GO entries. Moreover, nearly all of these entries are related to cell membrane functionalities including transmembrane transport. Accordingly, it can be speculated that when S. maltophilia is cultured with keratin as a substrate, keratin or some peptides derived from keratin can bind to transporters on the membrane and cross the cell membrane to enter the cell, providing growth substrates for bacterial metabolism. The keratin internalization is likely reliant on an uncharacterized transmembrane transport system in S. maltophilia. Additionally, the upregulation of these genes may also be stimulated by the entry of keratin into the cells and this positive feedback effect needs further investigation.

The same principles applied for GO enrichment analysis can also be used for Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of differentially expressed genes. The top-level pathway can be categorized into four main processes: cellular processes, genetic information processes, environmental information processes, and metabolic processes. These BPs can be subdivided into numerous second-level pathways. The number of differentially expressed genes in the second-level pathways was counted (Fig. 4), which showed that several pathways, including signal transduction, membrane transport, and amino acid metabolism, were significantly enriched. The second-level pathways also include several third-level pathways. Based on the KEGG enrichment results, with a threshold of p value ≤0.05, significantly enriched KEGG genes were selected (Fig. 5a). Gene expression differential analysis was further performed for each pathway (Fig. 5b). Significant upregulated expression was observed in the genes related to two-component systems in signal transduction, histidine metabolism, amino acid biosynthesis, biogenesis of cellular membranes, and ABC transport systems. The processes of upregulated gene expression imply that keratin needs to enter S. maltophilia cells through the membrane transport system. Furthermore, keratin can directly or indirectly promote the synthesis of keratinases, which can degrade keratin intracellularly or extracellularly by different enzyme subtypes.

To explore the proteins which were directly or indirectly binding with keratin, S-tag keratin monomers were used as bait in Co-IP coupled with mass spectroscopy. Total proteins by Co-IP from S. maltophilia were analyzed through silver staining and Western blotting (Fig. 6). Multiple protein bands suggest that many proteins bind to keratin monomers in S. maltophilia DHHJ (Fig. 6a). The Western blotting results show the presence of keratin as bait in the Co-IP (Fig. 6b), revealing efficient precipitation of S-tag keratin along with its interacting proteins in the assay. The sample underwent LC-MS/MS assay to identify protein components. As a result, 39 proteins were identified that directly or indirectly interact with keratin monomers in S. maltophilia DHHJ (Table 1). These proteins are overrepresented by molecular chaperones, peptide transport system proteins, and various proteases, consistent with the finding in RNA-seq experiment.

Based on the results from the above-mentioned RNA-seq and Co-IP experiments, four genes were selected for qRT-PCR, which included JJL50_15620 and JJL50_17955 (TonB-dependent receptors), JJL50_03260 (ABC transporter ATP-binding protein), and JJL50_20035 (ABC transporter substrate-binding protein). These genes showed a significant upregulation in response to keratin degradation in both transcriptome sequencing and immunoprecipitation analysis. When compared to commonly used nutrient sources (BPM), there was a high correlation between transcriptome and qRT-PCR data (R² = 0.88). In the presence of keratin as a nutrient source, the expression levels of the selected 4 genes exhibited an upward trend (Fig. 7).

Feathers, as one of the primary biological waste materials, have a global annual production of approximately 8.5 million tons [15]. In comparison with extensive investigations on the recycling of another natural bio-waste, cellulose [16, 17], there is relatively less attention paid to the utilization of nitrogenous biological resources represented by feather proteins. Especially, the mechanism of its degradation is unclear, directly hindering the improvement of degradation efficiency and further applications. Currently, research on feather degradation mainly focuses on the isolation and strain identification of degrading microorganisms, as well as the purification and screening of feather-degrading enzymes (keratinases). Currently, more than 50 related enzymes have been isolated from different microorganisms [18, 19]. These enzymes exhibit good feather protein degradation abilities inside cells, but they cannot completely degrade feathers in vitro [20]. Finding a single component enzyme that can completely break down feather proteins into biologically available amino acids and peptides is very difficult, even not impossible [5]. On the other hand, engineered bacteria expressing keratinases also lack the ability to efficiently degrade feathers [21, 22], indicating that the biological degradation of feather proteins may involve undisclosed mechanisms and complex degradation processes [23‒25]. The degradation mechanism of keratin is not yet fully understood. Researchers generally believe that this process consists of two steps: the breaking of disulfide bonds and the hydrolysis of peptide chains [20, 26]. These mechanistic studies primarily focus on the in vitro reaction of keratinases on keratin, with less emphasis on the overall degradation process in the biological context within microbial cells that degrade keratin. As a keratinase-producing microorganism, S. maltophilia DHHJ can grow well on a medium with feathers as the sole nutrient source (carbon source, nitrogen source, and energy source) and can completely degrade feathers or feather powder. Perplexingly, insoluble feathers or feather powder have much larger volumes compared to bacterial cells. How do they enter bacterial cells, and what is the transmembrane transport mechanism? How is the enzyme system in bacterial cells activated? The resolution of these questions is the key to efficient feather protein degradation and the potential utilization of feathers as a novel nitrogen source.

Our previous studies on S. maltophilia DHHJ indicate that the synergistic action between intracellular and extracellular enzymes achieves the biodegradation of feather proteins [27]. Feather proteins are composed of keratin monomers linked by disulfide bonds, and the monomers from different avian sources are highly conserved [28]. These monomers, encoded by the ker gene family, cross-link to form feather proteins through disulfide bonds [28‒31]. This study highlighted that the process of feather protein degradation not only depends on enzyme production and the secretion of extracellular enzymes but also requires the recognition of feather proteins or their degradation fragments by membrane-specific receptors. Feather keratins enter the cells through certain channels and are then enzymatically broken down into amino acids. It is worth noting that these internalized keratin monomers can induce expression of keratinase enzyme genes, potentially accelerating the turnover of feather proteins.

Transcriptome studies allow for the comprehensive exploration of gene function and structure, revealing molecular mechanisms underlying specific BPs and disease development [32]. Immunoprecipitation (Co-IP) analysis is an effective method for studying protein-protein interactions and can also identify unknown proteins interacting with specific proteins [33]. Both of these techniques have been applied in various metabolic research processes. In this study, by combining transcriptome sequencing with Co-IP results, we were able to discover differential genes involved in multiple metabolic pathways in the keratin degradation, which interact with keratin monomers. These proteins can be classified into three major categories: molecular chaperones, transport proteins on the cell membrane, and degradation-related enzymes.

Transport proteins are a major class of membrane proteins that mediate the exchange of chemical substances and signals between the inner and outer membranes of biological cells. This study identified three types of transport proteins that may be involved in the transmembrane transport of keratin (Fig. 6). The first type is the OmpA family protein, which is found in the outer membrane of prokaryotic cells and helps maintain outer membrane permeability [34]. The second type of transport protein is the TonB protein transport system, which is primarily present in Gram-negative bacteria. Bacteria use the TonB system to consume nutrients from the external environment. The TonB transport system consists of ExbB-ExbD and TonB proteins, with ExbB-ExbD located in the inner membrane and TonB in the periplasmic space. It provides energy for TonB-dependent outer membrane receptors through ATP hydrolysis, allowing the transport of various nutrients [35]. The third type of transport protein is the ABC transport protein family, which exists in both prokaryotic and eukaryotic organisms. In prokaryotes, ABC transport proteins are located on the cell membrane and use the energy generated by ATP hydrolysis to transport external nutrients across the membrane into the cytoplasm. They also participate in cellular nutrient intake, signal transduction, and other physiological processes [36‒38]. These three membrane transport systems, shown in Figure 8, may mediate cellular internalization of keratin monomers.

The second major category of proteins consists of various peptidases, with most of these peptidases belonging to the ATPases associated with diverse cellular activity (AAA+) proteases. The AAA+ protease system is the primary pathway for degrading intracellular proteins in prokaryotes and can degrade the majority of proteins within bacterial cells [39, 40]. AAA+ protease system is likely responsible for complete hydrolysis of keratin monomers.

Molecular chaperones constitute the third major category of proteins, among which chaperone protein DnaK belongs to the Hsp70 family. It is a highly conserved ATPase with a molecular weight of approximately 70 kDa, playing a crucial role in protein folding from scratch, transmembrane transport, and degradation of misfolded polypeptides [41]. Chaperone protein HtpG belongs to the Hsp90 family, which is involved in the folding and assembly of many signaling molecules in the Ras signaling pathway [42]. During the keratin degradation process, DnaK and HtpG may regulate the folding of keratin monomers to facilitate their cellular entry.

In summary, based on the data from transcriptome sequencing and Co-IP/mass spectrometry assays, we propose that the TonB and ABC transport systems potentially play a crucial role in feather keratin degradation by S. maltophilia DHHJ strain. Additionally, protein folding activity supported by molecular chaperones, such as DnaK and HtpG, may orchestrate keratin degradation.

An ethics statement was not required for this study type, and no human or animal subjects or materials were used.

The authors declare that they have no conflict of interest.

This work was supported by the National Natural Science Foundation of China under 31570106 and 31000989.

K.X. carried out RT-PCR and wrote the manuscript; X.X.S. and W.Z. carried out RNA-seq and Co-IP; Y.L.Z. prepared the figures; T.C. conducted reannotation of the gene; Z.J.C. conceived the study and approved the final version; F.H. and X.Q.Z. analyzed data; all authors have agreed to the final version of the manuscript.

All data generated or analyzed during this study are included in this article and its online supplementary material files. Further inquiries can be directed to the corresponding author.