Traceability and distribution of Neisseria meningitidis DNA in archived post mortem tissue samples from patients with systemic meningococcal disease

Meningococcal infections remain a major public health problem worldwide with a case fatality rate (CFR) of 7-11% in sporadic cases increasing to 20-52% in outbreak situations in Europe and USA [1‐3]. Approximately 30% of patients with systemic meningococcal disease (SMD) in Europe develop septic shock [2, 3]. Circulatory collapse is the primary cause of death owing to the combined effect of extreme vasodilation and septic cardiac failure resistant to treatment. The pathophysiology of SMD is closely associated with the ability of N. meningitidis to proliferate in the blood and subsequently invade the meninges, as documented by qPCR [2, 4‐11]. Fulminant meningococcal septicemia, the most feared clinical presentation, is characterized by rapid progression of septic shock, large petechiae and ecchymoses, disseminated intravascular coagulation (DIC) and renal and pulmonary failure [1, 2, 9, 12, 13]. Post mortem examinations reveal hemorrhagic adrenals in the majority of the patients, “shock”-kidneys often with multiple thrombi in glomeruli, normal or congested lungs sometimes with thrombi and occasionally inflammatory foci in epi- and myocardium [10, 13]. The findings are in line with those described by Ferguson and Chapman [14].

Patients presenting with distinct symptoms of meningitis without shock is the most common presentation and have a much better prognosis than those with shock [1‐3]. The number of meningococci and the LPS level in the plasma in these patients are low or undetectable [11] while the levels of meningococcal DNA, LPS and cytokines are 100- to 1000-fold higher in cerebrospinal fluid (CSF) than in blood [1, 2]. Death may be caused by brain edema and herniation of cerebellum.

The pathophysiology and outcome of meningococcal septic shock is closely associated with the plasma level of N. meningitidis lipopolysaccharides (LPS, endotoxin) and the circulating level of meningococcal DNA [4, 6, 11]. A rapid proliferation of N. meningitidis will result in huge amounts of LPS-containing material in plasma, and 95% of the patients with LPS levels in plasma above 10 endotoxin units (EU) /mL develop persistent shock with a detrimental activation of the innate immune system [1, 2, 9, 10, 12, 15, 16]. LPS, present in the membrane of the bacteria, are the most important but not the only meningococcal molecules that induce inflammation [17, 18].

A few studies have previously addressed detection and distribution of N. meningitidis DNA in FFPE human tissue [19, 20] but none has compared the results with FF tissue from the same post mortem examination. Recently, a porcine model of meningococcal septic shock, using the heat inactivated N. meningitidis, documented that large numbers of meningococci accumulated in the lungs, heart, liver, spleen and kidneys inducing a massive organ inflammation [21]. Similar results were found in three patients with lethal meningococcal septic shock by examining fresh frozen (FF) tissue from the same organs [21]. As an extension of this study, we aimed to detect and quantify N. meningitidis DNA i.e. copy numbers using qPCR in tissue samples from different organs obtained by post mortem examination. The tissue samples were formalin-fixed, paraffin-embedded (FFPE) and stored at room temperature (20 – 25 °C) for up to 28 years. A human endogenous DNA control was included to evaluate the effect of storage time on degradation of tissue samples. Furthermore, we compared the qPCR results obtained from FFPE tissue stored at 20 – 25 °C with FF tissue stored at −80 °C for up to six years.

In this study, we detected N. meningitidis DNA in FFPE and FF tissue from the lungs, heart, liver, kidneys and spleen in five patients dying of severe meningococcal shock and multiple organ failure. The highest concentration of N meningitidis DNA was found in the lungs and hearts. In one patient (No 4), with the highest levels of meningococcal DNA and LPS in plasma, N. meningitidis DNA was also detected in the brain FFPE tissue. In FF brain tissue from patient No 3 N. meningitidis DNA was also detected. Our results are in line with the current view that fulminant meningococcal septic shock and multiple organ failure is a compartmentalized infection with a massive proliferation of the bacteria in the circulation spreading to different large organs and the skin where they accumulate in large number. N. meningitidis can be detected in the CSF in shock patients but usually in much lower number as compared with patients with clinical symptoms of meningitis [4, 5, 11]. Our results are also in line with Fernández-Rodríguez et al. and Guarner et al. who detected N. meningitidis DNA in 81-100% of FFPE tissue from patients with meningococcal sudden deaths [19, 20]. Our previously published results also suggest that the majority of the nonviable bacteria could be disintegrated in various organs still exerting a powerful stimulus of the local innate immune system [21].

Brain tissue from two patients (6 and 7) dying of meningitis resulting in brain edema and cerebellar herniation, were without detectable N. meningitidis DNA in the brain and other organs. We know from previous studies that patients with meningitis as a group have high levels of LPS and inflammatory mediators as well as N. meningitidis DNA in CSF [1, 2, 5, 11, 29, 30]. In this study, the LPS level in CSF in one of two meningitis patients (No 7) was 4000 EU/mL, documenting a massive proliferation of meningococci in CSF. The storage time of tissue from brain and other organs, however, was 28 years at room temperature for both meningitis patients. We interpret our negative results in the brain tissue as a consequence of gradual degradation of DNA due to storage time (over 28 years) at room temperature and fixation in unbuffered formalin [31, 32]. Quantitative measurements of HBB DNA suggest that human DNA is degraded over time as indicated in Fig. 1. We assume that the same is the case for bacterial DNA in human tissue stored at room temperature.

In the two patients (No 6 and 7) dying of cerebellar herniation all extracranial organs examined, were N. meningitidis PCR negative. In addition to degradation of DNA as discussed above, the negative results could reflect low seeding of the different organs. Both patients had low levels of LPS in plasma (Table 1) indicating a low bacterial load in the circulation and possibly tissue concentrations of N. meningitidis below detection level (true negative results) [2, 5, 11]. Since few patients with distinct clinical meningococcal meningitis without shock die, we did not have any recent post mortem tissues to verify this hypothesis.

The patients with lethal pneumococcal disease serving as negative controls, had as expected no detectable N. meningitidis DNA. We did not quantify the amount of Streptococcus pneumoniae DNA in the different tissue in these two control patients, however, in a recently published study, comprising 11 patients with systemic pneumococcal infections, FF tissue from the large organs contained 10⁴ – 2 × 10⁶ pneumococci per gram tissue, the highest levels detected in the lungs [21].

Does the copy number of N. meningitidis in the different tissues represent bacterial components located primarily in the capillaries of the different organs or do they indicate meningococcal molecules, primarily LPS, which trigger tissue macrophages and induce local inflammation? A histological case study of one patient with lethal meningococcal sepsis suggested that meningococci mainly adhered to capillaries located in low flow regions in the infected organs [33]. Presumably they proliferate in the capillary vascular bed [33]. The capillary density is known to vary anatomically and functionally in different organs as well as within a single organ [34, 35]. How capillary density and volume may influence the distribution of N. meningitidis in different organs is not investigated in this study. However, we have previously found massive organ inflammation by using multiplex assay and quantified tumor necrosis factor (TNF), interleukin (IL)-1β, IL-6 and IL-8 in homogenized fresh frozen tissue samples from lungs, heart, liver, spleen, kidneys and adrenals from patients 3,4 and 5 [21]. These observations suggest that LPS and other neisserial molecules are recognized by tissue macrophages in the different organs, implying that the bacterial components are located in the tissues and trigger a local immune response varying from organ to organ.

Pre-analytical factors that may have an influence on the DNA analysis of FFPE tissue can be: postmortem interval (PMI), cold ischemia time (time between biospecimen removal from the body and its preservation), specimen size, fixative buffer (unbuffered formalin or neutral buffered formalin), fixative delivery method, fixative temperature and duration, block storage, section thickness and section storage, and methods of nucleic acids isolation [36‐41]. Several reports have shown that the FFPE method will lead to degradation of DNA over time due to the formalin fixation of tissue before paraffin embedding [38].

To monitor the quality of DNA extraction and to verify the presence of amplifiable DNA in our study, an endogenous DNA control gene HBB was quantified in FFPE and FF tissue (Fig1) [23, 42, 43]. Not surprisingly, the highest amount of HBB was found in the FF tissue, with Ct values of about 25 which indicates high amounts of high quality DNA as starting point for PCR amplification. When quantifying HBB in FFPE tissue, increasing Ct values, i.e. lower levels of intact DNA, were found with increasing storage time (storage time 28 years showed Ct values around 34) indicating degradation of DNA over time. In accordance with other studies this verifies that storage time of the FFPE tissue block has a major impact on the concentration of DNA found in the tissue [20, 31, 41, 44, 45]. Several reports show that amplification of DNA despite degradation of DNA may be possible [31, 41, 46]. The PCR products that are amplified in this study have amplicons under 100 base pairs. It is highly recommended to design the amplicons to be as short as possible when carrying out the qPCR of DNA from FFPE [47].

Storage time of FFPE tissue and deposit temperature are two parameters that may influence the traceability of specific DNA [20, 31, 41, 44, 45]. We conclude that meningococcal DNA can be detected in different tissue with the present qPCR assay after storage at room temperature for 11 years in meningococcal patients with shock and multiple organ failure. Storage for 28 years at room temperature gave negative results also in brain tissue with expected high concentration of meningococci in CSF on hospital admission. These results are supported by finding a significant positive association (r = 0.73) between storage time of tissue and amount of amplifiable endogenous control (HBB). Detectable levels of HBB DNA decline over the years, and suggest a degradation over time. The results are also in line with the findings of Fernández-Rodríguez et al. [20].

To evaluate the storage methods, we compared the levels of N. meningitidis DNA in fresh frozen (FF) tissue with FFPE tissue from three patients (3, 4 and 5) (Table 2 and Fig. 2). The tissue were 5, 2 and 6 years old, respectively. N. meningitidis DNA was detected in all tissue. In patient 3 and 5, the results showed a greater amount of N. meningitidis DNA in most FF tissue samples compared to FFPE, which is in line with previous studies [48, 49]. Patient 4 showed more similar levels of N. meningitidis DNA in the FFPE tissue and in FF tissue for most of the samples compared with patient 3 and 5. An explanation might be that the patient samples were only 2 years old and in good quality with less degradation of DNA.

This study suggests that N. meningitidis DNA is present in high concentrations in many of the major organs in meningococcal patients with shock and multiple organ failure [21]. N. meningitidis may induce a strong molecular inflammatory response in various tissues without distinct visible microscopical changes owing to the short duration of the infection before death [21].