Transcriptomic Responses of Acinetobacter harbinensis HITLi 7^T at Low Temperatures

Acinetobacter harbinensis HITLi 7^T, a novel species in the genus Acinetobacter, was isolated from Songhua River in Harbin, China. It is a psychrotolerant heterotrophic nitrification bacterium that grows in temperatures from 2 ℃ to 35 ℃ with an optimal growth temperature of 8 ℃- 20 ℃^[1]. The strain plays an important role in ammonium removal from source water and water sanitation improvement in cold areas. Biologically enhanced activated carbon (BEAC) filters immobilized with HITLi 7^T are used to treat source water in winter, exhibiting stronger ammonium removal capability and higher microbial stability during BEAC treatment at low temperatures^[2]. However, the low temperature adaptation mechanism of HITLi 7^T has not been clearly elucidated yet.

Temperature is a major environmental factor that affects bacterial growth. Low temperatures inhibit bacterial growth and alter physiological characteristics. Most research regarding bacterial adaptation to cold focuses on the cold shock response. Cold shock induces cold shock protein (Csp) production to reduce the harmful effect of temperature downshift^[3] and induces changes in expressed genes associated with metabolism, regulation, and fatty acid biosynthesis^[4].

In this study, we investigated the gene expression profiles of HITLi 7^T grown at a low temperature (2 ℃) and optimal temperature (20 ℃) by high throughput sequencing and transcriptomic analysis. We systematically assessed the change of expressed gene associated with adaptation to the low temperature conditions. Some differentially expressed genes (DEGs) were validated by RT-qPCR. This study facilitates to elucidate the molecular mechanism of HITLi 7^T adaptation to low temperatures and enrich our knowledge of the low temperature adaptation mechanism of the Acinetobacter genus.

HITLi 7^T was grown in liquid culture containing 0.382 g/L NH₄Cl, 2.0 g/L CH₃COONa, 0.05 g/L MgSO₄·7H₂O, 0.2 g/L K₂HPO₄, and 0.12 g/L NaCl with pH 7.0 at 20 ℃ for 24 h, re-suspended in the flesh culture, and incubated at 2 ℃ and 20 ℃ for 4 days. The experiment was conducted in triplicate for each temperature condition. The cultures were precipitated by centrifugation at 8, 000 rpm for 5 min and then washed with DEPC treated water. The cells were harvested and immediately frozen in liquid nitrogen.

Total RNA was extracted from cells using the Trizol reagent according to the manufacturer's instructions (Invitrogen). The quality and concentration of total RNA were checked by gel electrophoresis and with a spectrophotometer. Library construction and Illumina sequencing were performed at Huada Gene Technology Co., Ltd., Shenzhen, China. The library was constructed using 5 μg total RNA with TruSeq RNA Sample Prep Kit v2 from Illumina. The final library was detected by determining the average molecule length using Agilent 2100 bioanalyzer instrument (DNA 1000 Reagents) and quantifying concentration by qPCR (Taqman probe). The paired-end library was sequenced with the Illumina HiSeq 2000 system.

The raw reads were trimmed by removing low-quality reads (Q value < 20) and adapter sequences. The clean reads were assembled with Trinity assembler v2.0.6. The assembled sequences were clustered and redundancy was checked to obtain non-redundant transcripts with TGICL software v2.0.6. These transcripts were annotated to NT, NR, GO, COG, KEGG, SwissProt, and InterPro databases using Blastn, Blastx, Diamond, Blast2GO, and InterProScan5 software.

The RSEM package was used to measure gene expression level. Differential expression analysis was performed using DEseq2 method. The RPKM values were used for deriving the log fold changes. The probability threshold P-value ≤ 0.05 and the absolute value of log₂ (Fold Change) ≥ 1 were used as criteria for judging differentially expressed genes (DEGs). A principle component analysis (PCA) was conducted using DESeq2 to determine the overall reproducibility of RNA-Seq biological replicates. The DEGs were clustered using the pheatmap function in R software. All DEGs were classified and enriched in function and biological pathways by comparison on GO and KEGG databases using the phyper function in R software.

To validate the RNA-Seq results, 16 up-regulated genes were selected to perform RT-qPCR. The primers used in RT-qPCR were given in Supplementary Table S1 (available in www.besjournal.com). RT-qPCR was performed using StepOne (ABI, America) in a final volume of 20 μL containing 1 μL of cDNA, 10 pmol of each primer, 10 μL of 2 × SuperReal Color PreMix, 2 μL of 50 × ROX together with no template as negative control. A two-step reaction procedure was adopted: pre-denaturation at 95 ℃ for 15 min, denaturation at 95 ℃ for 10 s, annealing at 60 for 30 s, 40 cycles. All procedures were performed in triplicate. The relative gene expression levels were calculated using the 2^-ΔΔCT method with 16S rRNA as a reference gene.

The sequencing results of three parallel samples were similar; 13.73 Mb and 13.08 Mb clean reads were generated from treated and control samples, respectively. Removing partial overlapping and short sequences (< 300 bp in length) yielded 3, 104 unique genes with an average size of 1, 277 bp. These unique genes were annotated on the relevant databases. In total, 2, 535 unique genes were annotated on 7 databases, accounting for 81.6% of total unique genes (Table 1). KEGG analysis revealed that 56.6% of unique genes participated in metabolic processes and 20.7% were assigned to genetic information processing (Supplementary Figure S1A, available in www.besjournal.com). GO classification showed that 953 unique genes were assigned to 40 GO terms. More unique genes were attributed to metabolic and cell processes (Supplementary Figure S1B). COG annotation indicated that 1, 384 unique genes were clustered to 23 COG categories. The largest group is translation and ribosomal structure biogenesis, containing 258 genes (Supplementary Figure S1C).

Figure Supplementary Figure S1.  Function annotation statistics of unique genes.

In total, 393 DEGs were identified, of which 339 and 54 were up-and down-regulated, respectively, at 2 ℃, accounting for 86.3% and 13.7% of total DEGs, respectively (Figure 1). The results indicated that the larger number of DEGs induced up-regulation to adapt to low temperature conditions. GO function annotation showed that 289 DEGs were enriched in 28 GO terms (Supplementary Figure S2A, available in www.besjournal.com). Most DEGs were associated with metabolic, cellular, regulatory, and stimuli response processes. KEGG analysis indicated that 80 DEGs were enriched in 30 pathways. Most DEGs participated in metabolism pathways (Supplementary Figure S2B).

Figure 1.  Volcano plot of differentially expressed genes.

Figure Supplementary Figure S2.  GO and KEGG function annotation of differently expressed genes.

Of down-regulated genes, 39 were predicted to encode a hypothetical protein with unknown function and 19 were putatively involved in biological processes. The majority of hypothetical protein genes were down-regulated more than five times. Thus, further work is necessary to evaluate the possible role of these genes in low temperature adaptation. The gene Unigene1254_All encoding ClpV1 family T6SS ATPase and the gene Unigene1274_All encoding type Ⅵ secretion protein EvpB were putatively related to type Ⅵ secretion system (T6SS) and down-regulated (3.0-fold and 3.3-fold) in low temperature conditions, implying reduced T6SS-mediated intoxication toward cell growth at low temperatures. Some down-regulated genes were listed in Supplementary Table S2 available in www.besjournal.com.

The majority of up-regulated genes were involved in metabolic, regulatory, substance transport, translation, stimuli response, and cell division processes (Supplementary Table S3, available in www.besjournal.com), indicating that the adaptation of HITLi 7^T to low temperatures involves the extensive regulation of biological processes. A significant effect of low temperature on metabolism was observed. A large number of genes associated with carbohydrate and amino acid metabolism appeared to be up-regulated, implying the increase of the basal metabolic activity of HITLi 7^T at low temperatures and supply of sufficient substrate and energy for rapid growth at low temperature.

Carbohydrate metabolism provides energy and the carbon skeleton for cell growth; 39 genes related to carbohydrate metabolism appeared to be up-regulated. Seven genes involved in glycolysis and eight genes involved in citric acid cycle (TCA) were up-regulated, suggesting the enhancement of glycolysis and TCA at low temperatures. Additionally, several genes involved in pentose phosphate, propanoate, pyruvate, glycolate, carboxylic, and sugar metabolism were also up-regulated. Thus, we speculate that the increase of carbohydrate metabolism was one of the main mechanisms of HITLi 7^T low temperature adaptation.

Of genes related to amino acid metabolism and transporters, 42 were up-regulated, indicating the activation of amino acid metabolism at low temperatures. Some genes involved in lysine, leucine, arginine, histidine, glutamite, glutamate, and tryptophan synthesis were up-regulated, indicating that HITLi 7^T strived to accumulate and utilize these amino acids at low temperatures. Additionally, several genes related to lysine and leucine degradation and transport also appeared to be up-regulated, demonstrating the remolding of lysine and leucine metabolism, as well as uptake and utilization enhancement of lysine and leucine at low temperatures. Leucine is critical for production of branched-chain fatty acid^[5]. The up-regulation of genes (Unigene1324_All and Unigene536_All) encoding leucine synthesis and branched-chain amino acid transport may promote branched-chain fatty acid synthesis at low temperatures; the expression of these genes was validated by RT-qPCR (Figure 2).

Figure 2.  Validation of differently expressed genes by RT-qPCR.

Transcription regulation of HITLi 7^T in low temperature conditions was controlled by many global regulators, including RNA polymerase sigma factors, transcription regulator activators, temperature sensors, and AraC, Fur, IclR, Crp, LysR, ArsR, TetR, ArgP family transcription regulators. Sigma factor, as a bacterial transcription initiation factor, is responsible for regulating transcription initiation. Two RNA polymerase sigma factors encoding genes, Unigene1683_All (σ⁷⁰) and Unigene1454_All (σ³²), were up-regulated in low temperature conditions, implying the activation of transcription initiation. Additionally, σ³² is likely to positively regulate the expression of heat shock genes in HITLi 7^T at low temperatures due to the up-regulation of several heat shock genes (hscA, dnaJ, dnaK, htpG, clpB). Two transcription regulator activators, integration host factor subunit alpha (Unigene1333_All) and RNA polymerase-associated protein RapA (Unigene109_All), were 9.3-fold and 4.0-fold up-regulated in low temperature conditions, respectively, hinting the activation of transcription rate, frequency, and extent at low temperature. Additionally, a large number of transcription regulator encoding genes, such as AraC, Fur, IclR, Crp, LysR, ArsR, TetR, and ArgP family transcription regulators, were also up-regulated. These transcription regulators regulated multiple metabolic processes. Their up-regulation enhanced metabolic ability of the cell in low temperature conditions. The RT-qPCR results of the genes AraC (Unigene933_All), rpoH (Unigene1454_All), and RpoD (Unigene1683_All) were consistent with transcriptome analysis (Figure 2).

Of genes related to temperature and oxidative stress response, 15 were up-regulated. The Csps encoding gene (CL12.Contig1_All) exhibited an enhanced expression level of over 10-fold. Six hsps genes exhibited the enhanced expression (3.6-fold to 17.7-fold). Csps facilitate efficient translation at low temperatures by hampering the formation of the stable secondary mRNA structure acting as transcriptional activators or RNA chaperones. Hsps facilitates protein fold under stress by acting as molecular chaperones. Generally, hsps genes are up-regulated in increased temperatures^[6-7]. However, the opposite result was observed in our study. This is likely due to differences in molecular responses of different bacteria at low temperatures. The up-regulation of Csp (CL12.Contig1_All) and hsp (Unigene560_All and Unigene1443_All) were verified by RT-qPCR (Figure 2). Previous proteome analysis of HITLi 7^T also revealed the overexpression of two chaperone proteins in low temperature conditions ^[8]. These findings revealed that Csps and hsps played an important role in low temperature adaptation in HITLi 7^T by improving DNA binding and protein folding. The oxidative stress response genes, including catalase (Unigene902_All), superoxide dismutase (Unigene1524_All), flavoprotein (Unigene740_All), and alkyl hydroperoxide reductase subunit F (CL47.Contig2_All), were observed to be up-regulated, suggested that low temperature adaptation of HITLi 7^T also involved the oxidative stress response.

Low temperatures inhibit translation due to decreased ribosomal structural integrity and tRNA synthesis. Of genes associated with translation, 24 were up-regulated at low temperatures, including ribosome-associated genes, translation initiation factor IF2, IF1, translation elongation factor p, and tRNA synthesis- and transport related-genes. The up-regulation of these genes activated the translation of HITLi 7^T in low temperature conditions. Thus, we speculated that the translation activity of HITLi 7^T at a low temperature was enhanced by increasing ribosomal activity, mRNA longevity, translation initiation, and elongation as well as tRNA synthesis and transport. The RT-qPCR results of three genes (Unigene1084_All, Unigene751_All, and Unigene648_All) were consistent with transcriptome analysis (Figure 2) further revealed that translation level modulation may be an important process involved in cellular response to low temperatures.

To maintain nucleic acid structure and improve DNA replication and transcription at low temperatures, some genes related to DNA replication and recombination and RNA and DNA repair appeared to be up-regulated. Six DEAD-box RNA helicase encoding genes were significantly up-regulated (5.2-fold 37.8-fold), suggested that DEAD/DEAH box helicase played an important role in HITLi 7^T low temperature adaptation. DEAD-box proteins participate in transcription and translation as RNA helicase and chaperone and play a key role in cold tolerance in a multitude of organisms^[9-10]. One gene (CL16.Contig5_All) was validated by RT-qPCR (Figure 2). Several DNA gyrase and topoisomerase, exo- and endonucleases, DNA polymerase, DNA binding protein, DNA methyltransferase, DNA mismatch repair protein, and DNA recombination/repair protein genes involved in DNA replication and repair were induced in low temperature conditions. These findings indicated that the genes responsible for maintaining the activity and integrity of DNA and RNA are activated at low temperatures.

Of genes associated with protein, lipid, amino acid, organic, and ion transport and those associated with cell division proteins, 44 and 4, respectively, were up-regulated in low temperature conditions, indicating the improvement of substance transport and cell division at low temperatures. In addition, 66 hypothetical protein genes with uncharacterized function appeared to be up-regulated in low temperature conditions. Thus, more experiments are required to illuminate the function of these genes.

In conclusion, this study provides new insights into the low temperature adaptation of HITLi 7^T. The low-temperature adaptation of HITLi 7^T was elaborate and involved nearly every cellular process, mainly relying on basal metabolism regulation.