In human development, extensive physiological adaptation occurs during the neonatal period defined by the first four weeks of life (1). Among these, neonates have a developing immune system that is unique from adults and is associated with increased susceptibility to infections (2). The WHO reports that infections result in greater than 550,000 neonatal deaths each year (3). Among the most severe infectious conditions in neonates is sepsis, characterized by a blood infection with excessive inflammation and tissue pathology that occurs in an infant younger than 90 days old (4). Neonatal sepsis remains a major challenge globally, with the latest estimates suggestive of up to 5 million cases and about 800,000 deaths each year (5); as such, it results in significant mortality and neurodevelopmental impairment among survivors (6). GBS remains the most isolated pathogen in term infants, whereas Escherichia coli is more commonly responsible for mortality with preterm and low birth weight infants during early onset of sepsis (EOS) (7, 8).

Antibiotic therapy and general supportive care are the primary treatment options for neonatal sepsis (9). Common empirical antibiotic regimes with substantial geographical variations include the combinations of a penicillin and an aminoglycoside for EOS and a β-lactam antibiotic or a glycopeptide antibiotic plus an aminoglycoside for late onset of sepsis (LOS) (10). Emergence of resistant organisms, antibiotic-induced microbiome alterations, and downstream effects of high dose antibiotics on the developing immune system that lead to short- and long-term health outcomes in neonates are unresolved challenges in the treatment of neonatal sepsis. In particular, necrotizing enterocolitis (NEC) is a serious disease in preterm infants that is linked with antibiotics used in the early weeks of life. These drawbacks necessitate the discovery of new therapeutic strategies, and particularly those that reduce the dose or longevity of antibiotics (11). Alternative biological agents, such as bacteriophages, antibodies, anti-virulence agents, immune-modulating agents, and probiotics, are increasingly being explored as non-traditional alternatives to antibiotics (12).

Cytokines help to coordinate and resolve inflammation during sepsis (13). Though cytokine signaling can stimulate beneficial inflammatory mechanisms designed to clear microbial pathogens, excessive production of pro-inflammatory cytokines causes an uncontrolled inflammatory response commonly referred to as a "cytokine storm” that leads to tissue damage, multi-organ failure, and death during sepsis (14). Therapies targeting specific cytokines or cytokine pathways have been investigated as potential treatments for sepsis (15–17).

Interleukin (IL)-27 is a heterodimeric cytokine that belongs to the IL-6/12 cytokine family (18). IL-27 elicits pro-inflammatory activity by promoting T-cell differentiation to the Th1 lineage and inhibits the differentiation of Th2 cells (19, 20). It also suppresses Th17-cell differentiation and induces the production of the anti-inflammatory cytokine IL-10 (21). Reduced inflammatory cytokine production has been reported broadly for myeloid and lymphocyte subsets in response to IL-27 under a variety of acute and chronic stimuli (22–26). Thus, because of its diverse influences on the immune response, researchers have explored the therapeutic efficacy of IL-27 in various diseases, such as cancer and inflammatory diseases (27–30). We and others have reported that IL-27 contributes to enhanced susceptibility to bacterial infection, and inflammatory cytokine responses to gram-negative bacterial products are increased in neonates (31, 32). In our established neonatal murine sepsis model, we found that IL-27 signaling resulted in increased sepsis-associated morbidity and mortality compared to IL-27Rα-deficient neonatal mice that exhibited greater bacterial clearance with reduced systemic inflammation and improved glucose homeostasis (32, 33). Collectively, these findings are supported by evidence in a prospective study of human neonates suspected of infection in which IL-27 levels correlated with EONS disease prediction (34). In the work presented here, we have performed a preclinical study that investigated the efficacy of IL-27 antagonization in rescuing E. coli-induced neonatal sepsis in mice.

C57BL/6 (WT) and IL-27Rα-deficient (KO) mice were purchased from Jackson Laboratory (Bar Harbor, ME, USA) and maintained under specific pathogen-free conditions in the vivarium at West Virginia University Health Sciences Center. Mice were maintained on a 12-h light/dark cycle and were fed/watered ad libitum. Both male and female 4-day old pups were used for experimental infection.

Escherichia coli strain O1:K1:H7 was obtained from the ATCC (Manassis, VA) and grown in Luria broth from a single colony isolated on Tryptic Soy agar (TSA). To prepare infectious inoculums, the bacteria were enumerated as described previously (21). Neonatal pups (n = 3 or 4 per group [control, infected ± antibiotic ± anti-IL-27p28] per genotype [WT vs. KO]) at 4 days were inoculated subcutaneously in the scapular region with 50 µL of either PBS or E. coli O1:K1:H7 using a 28-gauge insulin needle as described previously (20, 21). The actual infectious dose for each experiment was determined by standard plate counts. Control and treated pups were identified by tail snips and tattoo markings, respectively. The weights of mice were recorded prior to infection and at 24 h post-infection just before euthanasia. For survival experiments, pups were observed for 5 days post-infection and weights were recorded every 6 h, accompanied by monitoring for signs of morbidity and mortality. This included poor weight gain, lack of spontaneous movements, absence of milk spot, and dehydration. All downstream experiments were associated with the same pre-identified mice. Organs were collected and placed in PBS or Trizol or 10% neutral buffered formalin solution for downstream analysis. Blood was deposited in tubes that contained 5 µL of 500 mM ethylenediamine tetraacetate acid (EDTA) and placed on ice. The bacterial burden in the blood and spleen were enumerated by serial dilution and standard plating on TSA. Blood glucose levels were measured using an AlphaTrack3 blood glucose monitoring system (Zoetis, MI, USA).

On the basis of previous studies by other research groups and initial standardization in our neonatal model, to neutralize IL-27, 10 µg anti-IL-27p28 monoclonal antibody (Biolegend, CA, USA) was subcutaneously injected into each pup 1 h prior to infection or 1–4 h following infection (35). Infected control mice were injected with an equivalent amount of IgG2 mouse isotype control antibody (Biolegend, CA, USA) prior to infection. For experiments in which pups were rescued with antibody alone, a lower infectious dose of 5–8×10⁴ CFUs (colony forming unit) was delivered. For antibiotic administration, a sub-lethal dose of gentamicin at which pups were unable to clear the bacterial burden was determined by administering two-fold dilutions of gentamicin in the range 2.5-0.125 µg/g at 1 h post-infection. We determined that for the infectious dose of 10⁵ CFUs/pup, the gentamicin concentration of 0.5 and 0.25 µg/g at 1 h post-infection is an ideal sub-lethal dose. For antibody and gentamicin combination studies we selected 0.5 µg/g gentamicin concentration at 2h post infection. In subsequent experiments, infected pups were rescued by administering 0.5 µg/g gentamicin alone or in combination with 10 µg anti-IL-27p28 antibody at 2 h post-infection. Infection outcomes and treatment efficiencies were assessed 24 h post infection and survival studies were conducted for 5 days. For experiments in which pups were treated with gentamicin alone or in combination with antibody, they were challenged with an infectious dose of 1–5×10⁵ CFUs/pup.

Spleens were thawed and homogenized in TriReagent®. RNA was isolated using the commercial product protocol. Briefly, the upper aqueous layer following phase separation was mixed with an equal volume of 75% ethanol and transferred to E.Z.N.A.® RNA isolation columns (Omega Biotek, Norcross, GA, USA). The manufacturer’s instructions were followed to complete tissue RNA isolation. iScript™ cDNA synthesis reagents (Bio-Rad, Hercules, CA, USA) were used to generate first strand cDNA according to the manufacturer's protocol. Real time cycling of reactions that included cDNA from the above preparation diluted 1:3 in nuclease-free water, gene-specific primer probe sets (Applied Biosystems, Foster City, CA, USA), and iQ™ Supermix (Bio-Rad) was performed in triplicate using a StepOnePlus™ (Applied Biosystems, Foster City, CA, USA) real time detection system. Cytokine specific amplification was normalized to that of actB as an internal reference gene and expressed as log₂ relative gene expression compared to control spleens using the formula 2^−ΔΔCt.

IL-6 and TNF-α serum levels were measured using multiplexed electrochemiluminescence U-plex reagents according to the manufacturer’s protocol (MesoScale Discovery [MSD], Rockville, MD, USA). Results were analyzed using MSD Discovery Workbench software (v4.0.13). Protein standards were assayed in parallel with samples.

In neonatal mice, IL-27p28 antibody treatment results in improved outcomes during E. coli-induced sepsis.

To evaluate the potential for IL-27 antagonization to improve outcomes in an experimental model of neonatal sepsis, we first investigated administration of antibodies alone as a monotherapy. We initially infected day 4 neonatal mice with an infectious dose of 5–8×10⁴ CFUs/pup. Anti-IL-27p28 antibody was administered 1 h prior to the E. coli infection (Fig. 1A). At 24 h post-infection there was significant weight loss observed in the infected pups that was improved in infected pups that received anti-IL-27p28 and insignificantly changed compared to uninfected pups (Fig. 1B). Infected pups that received anti-IL-27p28 were able to clear bacteria more efficiently from the blood and spleen (Fig. 1C-D). Infected pups had high bacterial burdens in the blood (mean CFUs of 3.15×10⁷/mL) and spleen (mean CFUs of 3.36×10⁴/spleen) at 24 h post-infection that were significantly reduced in IL-27p28 antibody received infected pups (mean CFUs of 5.34×10⁵/mL and 1.4×10⁴/spleen in the blood and spleen, respectively). It is known that IL-27 induces IL-10 production from a wide range of cell types including T-cell subtypes and macrophages (36, 37). In our model of neonatal sepsis induced by E. coli, we observed elevated gene expression levels of IL-10 in the spleens of infected pups (Fig. 1E) (33). Neutralization of IL-27 significantly reduced the gene expression of IL-10 in spleens at the time of infection relative to infected pups (Fig. 1E).

At 24 h post-infection, pups that received IL-27 antibody showed significantly decreased gene expression levels of cytokines IL-6 and TNF-α compared to infected neonates that did not receive treatment (Fig. 2A-B), although IL-6 and TNF-α protein levels were not different systemically in the serum between treatment groups (Fig. 2C-D).

Clinical application of IL-27 antagonization would most likely augment rather than replace antibiotic therapy. A preferred empiric regimen for EONS includes ampicillin and an aminoglycoside such as gentamicin (38, 39). Recommended doses of gentamicin range from 5-7.5 mg/kg (40). Before augmenting antibody neutralization with antibiotic treatment, we first evaluated the efficiency of genetic deletion of IL-27 signaling in controlling infection compared to antibiotic treatment. Both IL-27Ra KO and WT pups were infected with E. coli and after 1 h, pups were rescued with or without a single dose of gentamicin (5 mg/kg) (Fig. 3A). The median survival rate for infected WT pups was 32 h, which was significantly increased with antibiotic treatment to 68 h (Fig. 3B). Comparatively, infected KO pups had a median survival rate of 72 h without antibiotic treatment (Fig. 2B). Furthermore, KO pups were as efficient as gentamicin-treated WT pups in clearing bacteria from the blood and peripheral tissues (Fig. 3C-D).

To observe the potential added benefit of IL-27 neutralization, we needed to determine suboptimal rescue concentrations of gentamicin in our E. coli-induced neonatal sepsis model. We administered serially diluted concentrations of gentamicin to neonatal pups 1 h post-infection. The last concentration to provide full protection with CFUs in the blood and spleen at or below the limit of detection was 0.625 µg/g (0.625 mg/kg) (Fig. 3E). Infected pups that received 0.5 and 0.25 µg/g gentamicin showed detectable bacterial burdens in the blood and spleen, albeit reduced from infected pups in the absence of treatment, with moderate signs of sickness compared to infected pups (Fig. 3E). A lower concentration of 0.125 µg/g (0.125 mg/kg) body weight of gentamicin left infected pups unable to clear the blood and tissue bacterial burdens (Fig. 3E). These pups also exhibited signs of morbidity such as dehydration, poor feeding, and lethargy.

To evaluate the efficacy of antibody combined with antibiotic, pups were challenged with a high infectious dose of E. coli, 1–5×10⁵ CFUs/pup (Fig. 4A). At 24 h post-infection pups were severely morbid with a lack of mobility, hypoglycemia, and no milk spot visible as evidence of limited hydration and ability to nurse (Fig. 4B-C). A high bacterial burden of nearly 10⁸ CFUs/mL and 2×10⁵ CFUs/spleen was observed in the blood and spleens of infected pups, respectively (Fig. 4D-E). Pups that received a subclinical concentration of gentamicin (0.5 µg/g) at 2–4 h post-infection displayed significantly reduced glucose levels compared to uninfected pups, though they were not as morbid as infected pups that did not receive antibiotic (Fig. 4B-E). Spleen and blood burdens were lower than infected pups without treatment, although this did not reach statistical significance (Fig. 4D-E). Pups that were given the combination of gentamicin and IL-27p28 antibody were able to maintain blood glucose levels and active feeding was evidenced by a prominent milk spot and full stomach that was absent in the group treated with antibiotic alone (Fig. 4B-C). Bacterial proliferation in the blood and spleen was significantly restricted with the combination treatment relative to pups that received antibiotic alone (Fig. 4D-E). Pups that received gentamicin had an average of 1 log reduction of bacterial burdens from the blood and 0.2 log reduction from the spleen, respectively, relative to infected pups (Fig. 4D-E). A significant bacterial clearance was observed with the combination treatment; an average reduction of 1.7 and 2.25 log units was measured in the blood and spleen, respectively, compared to infected and untreated animals (Fig. 4D-E).

Serum cytokine levels of IL-6 and TNF-α were significantly lower in the pups that received the combination of gentamicin and IL-27 antibody and failed to increase comparable to infection with or without antibiotic alone (Fig. 5A-B).

To evaluate if more efficient bacterial clearance and other improved outcomes during neonatal sepsis at 24 h in response to IL-27p28 antibody and subclinical antibiotic therapy enhanced survival, we monitored pups through 120 h of infection. Infected pups without antibiotic struggled to maintain the body weight in the first day and all weighed less than the start of infection (Fig. 6A). Half of these animals succumbed to infection within the first day and the remaining by 48 h post-infection (Fig. 6B). Administration of 0.5 µg/g gentamicin promoted maintenance of body weight through 24 h, but 80% of the pups reduced weight thereafter (Fig. 6A). Conversely, 75% of the pups were able to gain weight throughout infection when given IL-27p28 neutralizing antibody and gentamicin (Fig. 6A). Supplementing gentamicin with IL-27p28 antibody significantly increased the 5-day survival rate from 10% to 41.6% (Fig. 6B).

In conjunction with high mortality rate, severe sepsis is also associated with tissue damage and organ failure (41). Histopathological analysis of H&E-stained liver sections from infected pups with and without gentamicin alone presented with necrotic regions and increased cell death (Fig. 7).

In the lungs, alveolar hemorrhage along with desquamation of epithelial cells into the alveolar spaces was prominent (Fig. 8). In contrast, treatment with IL-27p28 neutralization combined with gentamicin maintained liver and lung tissue architecture with minimal damage (Figs. 7–8).

Sepsis is responsible for significant morbidity and mortality in the neonatal population and therapeutic options for these infections are limited (42). Antibiotic therapy remains the only first line treatment arsenal beyond supportive care for neonatal sepsis, while major concerns are associated with evolving multidrug-resistant pathogens. In addition, long-term usage of antibiotics in neonates can cause necrotizing enterocolitis, a serious gastrointestinal condition (43). Recent advances in development of new neonatal sepsis treatments emphasize immunomodulatory approaches (44). Intravenous immunoglobulin (IVIg) and granulocyte colony-stimulating factor (G-CSF) are being investigated as potential adjuvant therapies (45). An important consideration is that neonates have distinct immune profiles and functions compared to adult immunity (46). Untrained immune cells with lower antimicrobial potential increase neonatal susceptibility to infection (47). With the aim to modulate the immature immune response of neonates to combat bacterial infections more efficiently, this study focused on the potential of IL-27 neutralization at the onset of neonatal infection.

We have established in previous studies using neonatal murine models that IL-27 compromises host control of bacteria consistent with elevated levels of inflammatory cytokines and increased mortality during sepsis (32, 33, 48). Conversely, mice that lack IL-27Rα, and cannot respond to the cytokine, exhibited improved pathogen clearance in peripheral tissue, reduced systemic inflammation, and limited mortality in septic neonatal mice (33). Using the same KO mouse model another scientific group observed improved bacterial clearance compared to wild-type controls during S. aureus infection, albeit in adult animals (49). The present investigation evaluated the therapeutic efficacy of IL-27p28 antibody to neutralize elevated levels of IL-27 cytokine present in septic neonatal animals. Consistent with our previous findings in mice that lack the ability to respond to IL-27, administration of IL-27p28 antibody to neonatal mice significantly improved bacterial clearance during E. coli induced sepsis. IL-27 signaling impairs bacterial clearance through multiple mechanisms. One such important mechanism is the negative regulation of the expression of V-ATPases which are recruited to late endosomes and lysosomes to promote acidification that is required to eliminate bacteria internalized through phagocytosis (50). Consistent with our findings, Bosmann and colleagues found that blocking of IL-27p28 during polymicrobial sepsis induced by cecal ligation and puncture resulted in improved bacterial clearance in blood and the peritoneal compartment along with improved survival rates and restricted cytokine release in adult mice (35). However, this work was done in adult animals. Understanding of the impact of IL-27 blockade with soluble neutralizing regents in neonates had not previously been explored.

The cytokine response is critical for establishing the severity and outcome of sepsis in newborns (51). In neonates, cytokines such as IL-6, TNF-α, and IL-1β are rapidly and significantly increased upon sepsis onset and characterized by a dysregulated inflammatory response which causes systemic inflammatory response syndrome (SIRS) that can contribute to organ damage (51, 52). A regulated cytokine response is essential for effective clearance of bacteria whereas overproduction of pro-inflammatory cytokines can lead to impaired function of immune cells (53). Previous studies from our group showed that IL-27 indirectly promotes an inflammatory cytokine response during neonatal sepsis, likely through impaired clearance of bacteria and microbial products. Accordingly, mice deficient in IL-27 signaling exhibited improved bacterial clearance and reduced levels of inflammatory cytokines (32, 33, 48, 54). In this study, pups that received IL-27p28 antibody displayed a regulated and balanced pro-inflammatory and anti-inflammatory cytokine expression levels in the spleen during infection consistent with improved bacterial clearance. A study of bacterial pneumonia caused by Streptococcus pneumoniae demonstrated that blocking signaling of IL-20 cytokine through its receptor improved bacterial clearance and decreased inflammatory reactions and tissue lesions (55). At the local tissue level there was a significant decrease in inflammatory cytokine gene expression in the spleen with IL-27 neutralization alone at 24 h post-infection in pups that received a moderate infectious dose. We did not observe significant changes in IL-6 and TNF-a serum cytokine levels. Though significant bacterial clearance was observed with neutralizing antibody treatment in the blood compared to infected pups, a single dose of neutralizing antibody alone is not sufficient to clear bacteria completely from the blood.

Experimental treatment failures and barriers to novel therapeutic approaches have left antibiotic therapy as the sole first line option to treat infections (56). This study for the first time, explored a method of supporting antibiotic treatment with an immune modulator for better prognosis of sepsis-related outcomes in neonates. Neonates are highly sensitive to medication and use of higher doses or longer duration of antibiotics in neonates causes organ specific damage (ototoxicity, nephrotoxicity and necrotizing enterocolitis), long-lasting dysbiosis and impairs immune development. Additionally, a study found that higher antibiotic exposure in the early life has been associated with an increased risk for several chronic conditions, such as respiratory allergies, atopic dermatitis, inflammatory bowel disease and attention deficit hyperactivity disorder (57, 58). IL-27 neutralization improved clinical outcomes of murine neonatal sepsis with lower doses of antibiotic, and as such, may represent a regimen that reduces the risk of NEC and other consequences of heavy antibiotics early in life. Our results demonstrated that IL-27p28 antibody treatment improved the efficacy of a subclinical dose of gentamicin and increased the survival rate of the neonatal septic mice. Pups that received the combination of gentamicin and IL-27p28 antibody had significantly improved bacterial clearance with a reduced inflammatory response, limited tissue damage, and improved outcomes during sepsis compared with pups that received antibiotic alone. Though we did not detect significant changes in the serum cytokine levels with IL-27 neutralization alone in the infected pups at 24 h post infection, the combination of antibiotic and antibody was effective in significantly decreasing levels of serum pro inflammatory cytokines IL-6 and TNF-α. Several studies have used antibodies designed to target specific bacterial toxins or virulence factors, enhancing the effectiveness of antibiotic treatment with consideration for the risk of antimicrobial resistance (59). However, our study focused on improving host response by suppressing a cytokine that increases susceptibility and opposes bacterial clearance during sepsis in neonates. Other researchers have explored the diagnostic accuracy of IL-27 and proposed the cytokine as a promising biomarker for neonatal sepsis (34, 60). The work presented here has extended those findings to validate IL-27 as a therapeutic target.

Limitations of the study include low the number of observations in each experimental group, which is a common obstacle with neonatal models in which a maximum 8–9 pups per litter limits 2–3 observations for each experimental group. This study used subcutaneous administration of either antibiotic or antibody as intravenous or intraperitoneal administration on day4 old pups is technically challenging and involves either backflow of the injecting material or increased risk of death in the pups. One of our future directions includes standardizing intravenous administration of neutralizing antibody to achieve maximum efficacy. This study has evaluated the efficacy of IL-27 neutralization in combination with one antibiotic, gentamicin, using Escherichia coli strain O1:K1:H7. This study can also be extended with different antibiotic choices (ampicillin, cefotaxime, vancomycin etc.) used to treat early onset of neonatal sepsis using other bacterial pathogens including clinical isolates.

Overall, our data suggest that the blockade of IL-27 signaling is beneficial to combat E. coli-induced sepsis in neonatal mice and can be augmented with antibiotic treatment. This strategy may have additional value with drug resistant bacteria and other clinical isolates associated with neonatal sepsis.