Genomics
Biomarkers
(reference)
Patients/animals
(reference)
Changes
Clinical relevance
(reference)
SNP
TLR1 [28]
1961 trauma patients
−7202G +742A/G
(Asn248Ser)
Association with increased mortality after traumatic injury and sepsis
14 septic patients and 30 healthy controls [30]
514 critically ill patients [31]
58 severely injured blunt trauma patients [32]
228 burn patients [33]
−260C → T
2. Association with increased susceptibility to sepsis [33]
3. Higher −260TT genotype frequency in ICU survivor patients [31]
IL-6、TLR4 and TNF-α [33]
228 burn patients
IL-6 -174 G → C
TLR4 +896A → G
TNF-α -308G → A
Association with increased risk for severe sepsis after burn injury
CpG
81 differentially methylated CpGs located in 64 genes [51]
3 septic and 3 non-septic babies
Protocadherin beta genes (PCDHB11/12/16/5/6/7/9) hypermethylated in newborns with sepsis. CCS-hypermethylated,
DEGS2-hypomethylated
Provide some novel insights into the role of DNA methylation in neonatal sepsis
4 Transcriptomics-Based Biomarkers
4.1 Gene Expression
The immune responses involved in sepsis are so complicated that the exact molecular mechanism remains to be fully elucidated [53]. The balance between pro-inflammatory responses and anti-inflammatory responses is closely related to the expression and regulation of relevant genes [54]. Hence, evaluating the key gene expression profiles by high throughput DNA chip may reveal the immune status of septic patients. Researchers have found specific changes of gene expression with microarray in certain organs and tissues of septic mice model, which including heart [55], liver, spleen [56], leucocytes [57] and so on.
Accordingly, Lukaszewsk recruited 92 ICU patients who had the risk of developing into sepsis [58]. The mRNA expression levels of IL-1β, IL-6, IL-8, IL-10, TNF-a, FasL and CCL2 in their blood leukocytes were measured on a daily basis by means of real-time reverse transcription PCR (RT-PCR), and analyzed with a nonlinear technique (neural network analysis). The data correctly predict the onset of sepsis in an average of 83.09 % of patient cases between 1 and 4 days before clinical diagnosis with high sensitivity and selectivity (91.43 and 80.20 %, respectively). Sutherland et al. [59] evaluated transcriptional profiles in circulating white blood cells of ICU sepsis patients, post-surgical patients and healthy controls with a microarray and multiplex tandem (MT)-PCR. A panel of 42-gene expression markers was identified, by which the prediction of sepsis within a mixed inflammatory population had an AUC between 86 and 92 %. Sepsis has a unique gene expression profile that is different from uninfected inflammation and becomes apparent prior to the clinical manifestations of sepsis for 0–48 h [60]. In that case, the specific gene expression profile, which may involve the function of innate immunity, cytokine receptors, T cell differentiation as well as the protein synthesis, may make a reference for early diagnosis of sepsis.
However, an important limitation of transcriptomics is that it only partially reflects the steady-state mRNA abundance, and the degree of mRNA abundance is influenced by multiple factors, and does not provide any direct information about gene end products (proteins), nor post-translational modifiers of protein function [61]. When it comes to the sample used as RNA source, there is a contradiction. For the whole blood approach, it may be difficult to interpret the confounded data of RNA for the reason of the heterogeneity among blood cell populations. As for the cell-specific approach, there is a possibility to miss relevant expression information from other cells due to the complexity of clinical sepsis [61]. It highlights the necessity of linking theory to clinical practice.
4.2 miRNAs
MicroRNAs (miRNAs) is a class of short RNAs with 18–25 nucleotides in length which regulate gene expression in a post-transcriptional manner via sequence-specific interaction with target sites in mRNA [62], associated with various physiological and pathological processes. The levels of miRNA in serum and plasma are consistent among individuals of the same species, resistant to RNase A digestion, and stable even after the freeze-and-thaw and a long term of storage [63, 64]. The stability of miRNA makes it a potentially useful candidate for diagnostic and other clinical applications. Although the source of circulating miRNA is still unclear, it has been proved that there is a link among a range of diseases, such as circulating miRNA and cancer [65, 66], trauma [67, 68], acute pancreatitis [69], and hepatitis [70].
When it comes to sepsis, by using genome-wide miRNA profiling with microarray in peripheral blood leukocytes and quantitative RT-PCR, Vasilescu [71] found that miR-150 levels were significantly reduced in both leukocytes and plasma of sepsis patients and had a negative correlation with the level of disease severity measured by the Sequential Organ Failure Assessment (SOFA) score, which made it a biomarker of early sepsis. Similarly, Zeng [72, 73] investigated the levels of miR-150 and miR-143 in peripheral blood leukocytes in sepsis patients with RT-PCR, and found that the expression levels of miR-150 and miR-143 were significantly decreased in sepsis patients and could reflect the severity of sepsis in certain degree, which not only made it a marker to reflect the situation of inflammatory response, but also made it a prognostic marker in sepsis. Recently, higher serum miR-133a levels were found among sepsis patients in ICU [74]. As they were significantly correlated with disease severity, classical markers of inflammation and bacterial infection, as well as organ failure, high miR-133a levels were considered as independent biomarkers for unfavorable prognosis of critically ill patients.
However, given that the pathophysiological process of sepsis involves a variety of tissues and organs, a simple screen for miRNA differentially expressed in leukocytes may omit those secreted by other cells. Wang et al. [75] used genome-wide microarray to identify differential serum miRNAs in survival and non-survival sepsis patients, and then further validated the differential expressions of miR-297 and miR-574-5p by RT-PCR in a larger group. The serum miR-574-5p together with sepsis stage and Sepsis-Related Organ Failure Assessment scores has a better predictive capability for the death of sepsis patients. In addition, serum miR-146a and miR-223 were found significantly reduced in septic patients compared with SIRS patients and healthy controls which might serve as new biomarkers for sepsis with high specificity and sensitivity [76]. Due to our knowledge on serum miRNAs is still at a primary stage, the expression level of circulating miRNAs at different stages of sepsis and their potential correlation with injured organs need further investigation.
To sum up, from the point of view of gene transcription, miRNA may undertake the task of diagnosing sepsis in an early stage and evaluating the prognosis, as well as becoming the new target for sepsis therapy.
4.3 Long Non-coding RNAs (LncRNAs)
As discussed above, epigenetic factors not only include histone modifications and DNA methylation, but also contain non-coding RNAs(ncRNAs), which have diverse size and can be generated from intergenic regions, introns, or enhancers [45]. LncRNAs are transcripts longer than 200 nucleotides and lack protein-coding capacity. Peng et al. [77] first discovered the widespread differential expression of lncRNAs in response to severe acute respiratory syndrome coronavirus (SARS-CoV) virus infection. Accordingly, there is a possible link between lncRNAs and the host defense response against infection. LncRNA has the potential to become new class of biomarkers and new therapeutic target for infectious diseases. However, as the functions of lncRNAs remain largely unexplored, there is a need for future studies on their regulatory role in infection .
All the transcriptomics-based biomarkers mentioned above are outlined in Table 2.
Table 2
Potential transcriptomic biomarkers for sepsis
Transcriptomic | Biomarkers (reference) | Patients/animals (reference) | Changes | Clinical relevance (reference) |
|---|---|---|---|---|
Gene expression | A panel of 42 sepsis gene expression markers [60] | Mixed inflammation group (28 sepsis and 38 post-surgical patients in ICU), and 20 healthy controls | NA | A novel molecular biomarker test has the capacity for early detection of sepsis via the monitoring of patients |
IL-1β, IL-6, IL-8, IL-10, TNF-a, FasL and CCL2 mRNA expression [59] | 92 ICU patients | NA | Provide a generic indicator of sepsis and help its early diagnosis | |
miRNAs | 17 sepsis patients and 32 healthy controls [72] 40 sepsis patients,20 SIRS patients, and 20 health controls [73] | ↓ | The miR-150 levels in both leukocytes and plasma correlate with the aggressiveness and prognosis of sepsis and can be used as a marker of early diagnosis | |
miR-133a [74] | 223 critically ill patients(138 with sepsis and 85 without sepsis) and 76 healthy controls | ↑ | High miR-133a levels were associated with the severity of disease and predicted an unfavorable outcome of critically ill patients | |
miR-143 [73] | 40 sepsis patients, 20 SIRS patients and 20 healthy controls | ↓ | The expression level of miR-143 may be a marker for judging the severity of sepsis and its prognosis | |
miR-146a [76]; miR-223 [76] | 50 sepsis patients, 30 SIRS patients and 20 healthy controls | ↓ ↓ | Serum microRNAs might be used as biomarkers for early diagnosis and reflecting severity of sepsis | |
miR-574-5p [75] | 12 surviving and 12 nonsurviving sepsis patients for microarray scan; 118 sepsis patients for validated by qRT-PCR | ↓ | The miR-574-5p combined with SOFA scores and sepsis stage provides a prognostic predictor of sepsis patients |
5 Proteomics-Based Biomarkers
Proteome is the complete set of proteins that can be expressed by the genetic material of an organism. Proteomics is the analysis of the expression, localizations, functions, and interactions of proteomes. Compared to other immunologic tests, proteomics is a novel method with advantages of high throughput, high sensitivity and specificity. The development of proteomics has allowed for a better understanding of the molecular bases concerning the identification of cell signaling, modifying protein, post-translation modification pathway, as well as the characterization of specific biological markers [78].
Proteomics has irreplaceable clinical significance and an expansive application prospect in studies of sepsis biomarkers . In a rabbit sepsis model by intravenous injection of Pseudomonas aeruginosa at 24 h after scald, 11 discrepant expression proteins from lymphocyte were found by matrix-assisted laser desorption/ionization time of flight mass spectrometry(MALDI-TOF MS). They are related with the folding, assembling, transportation and degradation of proteins, signal transmission, inflammation, immunization, energy metabolism, the proliferation, differentiation and apoptosis of cells [79]. In a recent research, 41 differential expressed proteins in the neutrophils from Acinetobacter baumannii sepsis rats were identified using two-dimensional electrophoresis and mass spectrometry [80]. They included antioxidant proteins, signaling proteins, cytoskeleton and regulatory proteins, energy metabolism and protease protein, which may play a key role in such kind of sepsis and provide potential clues in early diagnosis and treatment of sepsis.
In clinic, YKL-40 was identified with proteomics analysis on a significantly higher expression level in serum samples from sepsis patients and considered as a possible biomarker of sepsis [81]. Paugam-Burtz et al. [82] used plasma profiling coupling proteinchip array with surface-enhanced laser desorption ionization time-of-fly mass spectrometry (SELDI-TOF MS) to analyze the plasma of postoperative patients, and found that a combination of five plasma protein peaks may have potential as diagnostic biomarkers of postoperative sepsis in patients undergoing liver transplantation. Even so, these proteins remain to be identified and validated in more clinical trials.
6 Metabolomics-Based Biomarkers
Although many potential sepsis biomarkers have been revealed by genomics, transcriptomics, and proteomics, the changes of cellular metabolism in sepsis should be paid attention. Metabolomics is an emerging omics technology following genomics and proteomics and focuses on the metabolic products with a molecular weight less than 1000 kD under the physiological or pathological status. It can analyze the biochemical events of cells, tissues or organs and evaluate the disease and its severity. The research methods of metabolomics mainly include nuclear magnetic resonance (NMR), gas chromatography/mass spectra (GC/MS), high performance liquid chromatography/mass spectra (HPLC/MS).
The development of sepsis involves the reactions of multiple systems on various levels, which has a significant influence on the expression levels and activities of metabolic enzymes. And by detecting the concentration and ratio changes of those metabolites involved, a better understanding of condition and prognosis of sepsis may be achieved at an early stage [83]. Metabolic profile of the serum from septic rats with cecal ligation and puncture was achieved with the help of NMR and LC/MS [83]. In the septic rats, especially the non-survivors, many free fatty acids showed a lower level which may be consumed greatly for energy supply in sepsis and may be related with the prognosis of sepsis. Moreover, there was a rise of some polyunsaturated fatty acids in the very group, which may have a relationship with the increased anti-inflammatory effect. Based on the metabolic profile analysis, a model for outcome predication was built with high sensitivity and specificity, which provided a novel method for sepsis prognosis judgment. NMR-based metabolic profiling revealed the difference of metabolites of energy metabolism and inflammation in lung tissue, bronchoalveolar lavage (BAL) fluid, and serum samples between the septic rat and the control rat [84]. In septic rats, creatine concentration increased in all the three types of samples, whereas alanine and phosphoethanolamine concentrations increased only both in lung tissue and in serum. Myoinositol increased in lung tissue but decreased in BAL fluid. In addition, acetoacetate increased whereas formate decreased in serum. And with the construction of a predictive model for diagnosis of sepsis using partial least-squares discriminant analysis, the preliminary goal of sepsis diagnosis was achieved.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
