Masafumi Nukina

Antibodies to gangliosides and galactocerebroside in patients with Guillain–Barré syndrome with preceding Campylobacter jejuni and other identical infections

The relationship between preceding infections and antibodies to glycolipids was investigated in 205 Japanese patients with Guillain-Barré syndrome (GBS). Serological evidence of recent Campylobacter jejuni (C. jejuni) infection was found in 45% of the patients, compared with 1% in healthy controls. In contrast, recent infection of cytomegalovirus (CMV), Mycoplasma pneumoniae (M. pneumoniae) and Epstein-Barr virus (EBV) was detected in only 5%, 2% and none of the patients, respectively. C. jejuni-associated GBS was more frequent in early spring than in other seasons. All stool specimens positive for C. jejuni isolation were obtained within 10 days after the onset of GBS symptoms. Of 13 C. jejuni isolates from GBS patients, 10 (77%) belonged to Penner serotype 19 (heat-stable, HS-19). Elevated titers of anti-GM1 antibody were found in 8 (80%) of 10 GBS patients whose C. jejuni isolates belonged to HS-19 and in none of those infected with non-HS-19 C. jejuni (P = 0.04), and in 49% of 92 patients with C. jejuni infection and 25% of patients without infection of C. jejuni, CMV, EBV, or M. pneumoniae (P = 0.0007). The frequencies of elevated antibody titers to GD1a, GD1b and GQ1b were also significantly higher in GBS patients associated with C. jejuni than those not associated with C. jejuni, CMV, EBV, and M. pneumoniae. GBS in Japan seems to be associated more frequently with C. jejuni and less frequently with CMV than in Europe and North America.

Anti-GalNAc-GD1a antibody–associated Guillain-Barré syndrome with a predominantly distal weakness without cranial nerve impairment and sensory disturbance

The serum antibodies to N‐acetylgalactosaminyl GD1a (GalNAc‐GD1a) and other gangliosides as well as to Campylobacter jejuni were determined in 147 patients with Guillain‐Barré syndrome (GBS). We found a distinctive clinical pattern in patients with anti‐GalNAc‐GD1a antibodies compared with those without the antibodies, that is, lack of cranial nerve involvement (87% versus 38%), distal‐dominant weakness (80% versus 25%), and no sensory disturbance (73% versus 22%). The frequency of distal‐dominant weakness was significantly higher in patients with both C jejuni infection and anti‐GalNAc‐GD1a positivity (100%) than in C jejuni–negative/anti‐GalNAc‐GD1a–positive (25%), C jejuni–positive/anti‐GalNAc‐GD1a–negative (32%) and C jejuni–negative/anti‐GalNAc‐GD1a–negative patients (20%). Lack of cranial nerve involvement and sensory disturbance were found in most C jejuni–positive/anti‐GalNAc‐GD1a–positive and C jejuni–negative/anti‐GalNAc‐GD1a–positive patients, but not in C jejuni–positive/anti‐GalNAc‐GD1a–negative and C jejuni–negative/anti‐GalNAc‐GD1a–negative patients. Although the anti‐GM1–positive/anti‐GalNAc‐GD1a–negative patients mostly (75%) lacked cranial nerve involvement, distal‐dominant weakness (38%) and lack of sensory disturbance (13%) were infrequent. These results may indicate that (1) the combination of C jejuni infection and anti‐GalNAc‐GD1a antibodies, but not anti‐GalNAc‐GD1a, anti‐GM1, or C jejuni infection alone, is associated with a predominantly distal weakness, (2) the presence of anti‐GalNAc‐GD1a, rather than C jejuni infection or anti‐GM1 antibody, is associated with a lack of sensory disturbance, (3) both anti‐GalNAc‐GD1a and anti‐GM1 antibodies are independently associated with a lack of cranial nerve impairment. Ann Neurol 1999;45:758–768

Development and evaluation of a loop-mediated isothermal amplification assay for rapid and simple detection of Campylobacter jejuni and Campylobacter coli

We developed a loop-mediated isothermal amplification (LAMP) assay for the rapid and simple detection of Campylobacter jejuni and Campylobacter coli. The assay provides a specific LAMP product for each of these two species. The assay correctly identified 65 C. jejuni and 45 C. coli strains, but not 75 non-C. jejuni/coli strains. The sensitivity of the LAMP assay for C. jejuni and C. coli in spiked human stool specimens was 5.6 x 10(3) c.f.u. g(-1) (1.4 c.f.u. per test tube) and 4.8 x 10(3) c.f.u. g(-1) (1.2 c.f.u. per test tube), respectively. When 90 stool specimens from patients with diarrhoea were tested by LAMP and direct plating, the LAMP results showed 81.3 % sensitivity and 96.6 % specificity compared to isolation of C. jejuni and C. coli by direct plating. Further, the LAMP assay required less than 2 h for detection of C. jejuni and C. coli in stool specimens. This LAMP assay is a rapid and simple tool for the detection of C. jejuni and C. coli and will be useful in facilitating the early diagnosis of food poisoning incidents caused by these organisms.

Characterization of Legionella pneumophila isolates from patients in Japan according to serogroups, monoclonal antibody subgroups and sequence types

We collected 86 unrelated clinical Legionella pneumophila strains that were isolated in Japan during the period 1980-2008. Most (80.2%) belonged to serogroup 1, followed by serogroups 5, 3 and 2. Interestingly, the patients with L. pneumophila serogroup 1 had a significantly higher male-to-female ratio (12.4) than the patients with other L. pneumophila serogroups (2.0) (OR, 10.5; 95% CI, 2.5-44.5). When the serogroup 1 strains were analysed by monoclonal antibody (mAb) typing, the most prevalent subgroup was Benidorm (34.9% of all isolates). Moreover, 79.7% of the serogroup 1 isolates were bound by mAb 3/1, which recognizes the virulence-associated epitope. When all 86 isolates were subjected to sequence-based typing (SBT) using seven loci, they could be divided into 53 sequence types (STs). The ST with the most isolates (seven) was ST1, to which most isolates from patients and environments around the world belong. However, six of the seven ST1 isolates were isolated before 1994. Other major STs were ST306 (n=6), ST120 (n=5) and ST138 (n=5). All ST306 and ST138 isolates, except for one isolate (ST306), were suspected or confirmed to be derived from bath water, which suggests that these strains prefer bath habitats. The sources of all ST1 and ST120 isolates remain unclear. By combining the SBT and mAb data, the 86 isolates could be divided into 59 types (discrimination index, 0.984). This confirms the usefulness of this combination in epidemiological studies.

Characterization of Campylobacter jejuni isolates from patients with Guillain-Barré syndrome

Campylobacter jejuni is a major pathogen preceding Guillain-Barré syndrome (GBS), and most C. jejuni isolates from GBS patients belong to Penner serotype 19 (heat-stable; HS-19). We analyzed sixteen independent clinical isolates from GBS patients, twelve of which belonged to HS-19, three to HS-2, and one to HS-4, using PCR-based RFLP analysis of a flagellin-A (flaA) gene. Two isolates from patients with Miller Fisher syndrome (MFS), and 27 from patients with uncomplicated enteritis were also examined. All HS-19 isolates, regardless of GBS, showed an identical pattern (Cj-1) by RFLP typing and were distinguishable from those of the other Penner serogroups. In contrast, HS-2 and HS-4 isolates were divided into several different RFLP groups, suggesting HS-19 strains are genetically distinctive among C. jejuni isolates. A DNA fingerprinting method also failed to detect any specific band pattern for GBS-related C. jejuni isolates. We examined relationships among anti-GM1 antibody titres in the sera of GBS patients, clinical forms of GBS, serotype of C. jejuni, and the presence of GM1-like structures in lipopolysaccharide (LPS) components from C. jejuni isolates by immunoblotting. HS-19 related GBS was significantly associated with elevated anti-GM1 antibody titers in the sera of the patients, but not associated with any clinical pattern of GBS. No significant correlations were found between anti-GM1 antibody and the pattern of disease, or between GBS-related C. jejuni strains and the presence of GM1-like structures. HS-19 strains seem to be unique among C. jejuni isolates, and HS-19-related GBS may provide an excellent model for clarification of the pathogenesis of GBS.

Campylobacter jejuni isolates from Japanese patients with Guillain-Barré syndrome.

Serologic evidence of recent Campylobacter jejuni infection was found in 92 (45%) of 205 Japanese patients with Guillain-Barré syndrome (GBS), and 49% of those 92 patients also had antibodies to GM1. Sixteen independent clinical isolates from GBS patients were serotyped: 12 belonged to Penners heat-stable (HS) O serotype HS-19, 3 to HS-2, and 1 to HS-4. Of the patients whose C. jejuni isolates belonged to HS-19, 80% had elevated anti-GM1 antibodies. Although the correlation was significant between C. jejuni and GM1 antibody, anti-GM1 also was detected in 25% of patients without C. jejuni infection. Polymerase chain reaction-based restriction fragment length polymorphism analysis of an flaA gene showed that all HS-19 isolates, regardless of a GBS association, had an identical and distinguishable pattern, Cj-1, suggesting that HS-19:Cj-1 isolates are distinctive among C. jejuni isolates. Lectin typing showed that all GBS-associated HS-19 isolates contained terminal beta-N-acetylglucosamine residues on their cell surface, but HS-19 isolates from patients with enteritis did not.

Development of a loop-mediated isothermal amplification assay for sensitive and rapid detection of Campylobacter fetus

We developed a sensitive and rapid loop-mediated isothermal amplification (LAMP) assay for detection of Campylobacter fetus. This assay provides simpler and more rapid detection of C. fetus than conventional biochemical and PCR assays. The assay correctly detected 60 C. fetus strains but not 55 non-fetusCampylobacter and 30 non-Campylobacter strains. The sensitivity of the LAMP assay in pure cultures and in a spiked bovine liver specimen was 10-fold more sensitive than that of the conventional PCR assay. The sensitivity of the LAMP assay in a spiked bovine vaginal mucus specimen was similar to that of the conventional PCR assay. The assay was markedly faster, requiring less than 40min for detection of C. fetus in a single colony on blood agar and 80min in spiked bovine specimens from the beginning of DNA extraction to final determination. Our LAMP assay is a simple and practical tool for detection of C. fetus regardless of subspecies.

Synergistic effect of HLA class II loci and cytokine gene polymorphisms on the risk of gastric cancer in Japanese patients with Helicobacter pylori infection

It has been reported that polymorphisms of human leukocyte antigen (HLA) genes and several cytokine genes are associated with an increased risk of developing gastric cancer (GC). However, the results of studies from different geographic regions, ethnic groups and study groups are inconsistent. The aim of this study was to evaluate the influence of H. pylori infection and host genetic factors on GC susceptibility in Japanese patients with GC. We analyzed genotypes for HLA class I and II, tumor necrosis factor α, interleukin (IL)‐1β, IL‐1 receptor, IL‐4, IL‐4Rα and IL‐10 in 330 H. pylori‐infected noncardia patients with GC and 190 H. pylori‐infected nonulcer dyspeptic controls. Haplotype analyses indicated that the frequencies of the HLA DRB1*0405 and DQB1*0401 alleles were increased in the patients with intestinal‐type GC when compared with controls (both DRB1*0405 and DQB1*0401: p = 0.015, OR = 1.57, 95% CI = 1.09–2.26), but the changes were not statistically significant after correction for multiple comparisons. None of the cytokine gene polymorphisms were associated with GC susceptibility, whether patients with GC were analyzed as a group according to the histological subtype. Of interest was the comparison of controls and patients with intestinal‐type GC. The frequency of an IL‐10‐592AA homozygote showing concomitant carriage of the HLA DRB1*0405‐DQB1*0401 haplotype was significantly higher in patients with intestinal‐type GC (χ2 = 6.369, p = 0.0116, pc = 0.0464, OR = 2.43, 95% CI = 1.21–4.48). Our results suggest that the HLA class II and IL‐10‐592A/C polymorphisms synergistically affect the susceptibility to GC development of H. pylori‐infected individuals in the Japanese population.

Three patients with ophthalmoplegia associated with Campylobacter jejuni.

Cranial polyneuropathy is idiopathic in most patients. Idiopathic cranial polyneuropathy is an acute postinfectious syndrome, along with Guillain-Barré syndrome and Miller Fisher syndrome, in which the common preceding pathogen is Campylobacter jejuni. Serum anti-GQ1b antibodies are elevated in Miller Fisher syndrome and Guillain-Barré syndrome with ophthalmoplegia. Three patients with idiopathic cranial polyneuropathy with predominant ocular involvement are presented. C. jejuni isolated from stool specimens belonged to Penner serotypes O:4, O:23, and O:33. Serum anti-GQ1b antibodies were elevated in all patients but demonstrated rapid reduction concomitant with clinical recovery. All patients recovered completely. Because both preceding C. jejuni infection and elevated anti-GQ1b antibodies decreasing with time were seen in all patients, the pathogenesis of idiopathic cranial polyneuropathy with ophthalmoplegia may be similar to that of Miller Fisher syndrome.

Purification of Clostridium botulinum type G progenitor toxin

Clostridium botulinum type G cultured for 6 days at 30 degrees C in proteose peptone-yeast extract-glucose medium produced toxin of 1.3 x 10(4) LD50/ml. The toxin was precipitated at pH 4.0, extracted with 0.2 M phosphate buffer, pH 6.0, and activated with trypsin. Sonic treatment and trypsinization of the residual precipitate released additional toxin, the toxicity of which corresponded to that detected in whole culture. Activated toxin obtained from the first extract and that from the residual precipitate were combined and purified by salting out, acid precipitation, gel filtration on Sephadex G-200, chromatography on SP-Sephadex, and a second gel filtration on Sephadex G-200. The yield of purified toxin from 10 liters of culture was 22.9 mg an 1.1 X 10(8) mouse ip LD50 with a specific toxicity of 3.0 X 10(7) mouse ip LD50/mg nitrogen. The molecular weight of the toxin was about 500,000, corresponding to that of L toxin of the other types. No M nor LL toxin was detected.