The Epidemiology of Meningococcal Disease and Carriage, Genotypic Characteristics and Antibiotic Resistance of Neisseria meningitidis Isolates in Zhejiang Province, China, 2011–2021

Neisseria meningitidis (Nm) remains a worldwide leading cause of epidemic meningitis. During 2011–July 2021, 55 meningococcal disease (MD) cases were reported with a case fatality rate of 5.45% in Zhejiang Province, China. The median age was 7 years. The annual incidence was 0.0017–0.0183 per 100,000 population. The highest age-specific incidence was observed in the group younger than 1 year. Serogroup was identified in 30 laboratory-confirmed MD cases, and MenB was most predominant. MenB was mainly observed in two age groups: younger than 5 and older than 35 years. MenB incidence was significantly increasing from 0.0018 per 100,000 in 2013 to 0.0070 per 100,000 in 2019. During 2015–2020, 17 positive samples were detected from 2,827 throat swabs from healthy population, of which 70.59% was MenB. Twenty multilocus sequence typing sequence types (STs) containing eight newly assigned STs (ST15881–ST15888) were determined in all Nm isolates. Either in MD cases or in healthy population, MenB CC ST-4821 was the predominant ST. It was worth noting that two MenY CC ST-23 cases occurred in 2019 and 2021, respectively. MenY CC ST-23 MD cases increased gradually in China. Phylogeny results based on genome sequencing indicated that Chinese MenW CC ST-11 isolates were genetically linked and grouped together with Japanese isolates, separated from MenW CC ST-11 isolates from Saudi Arabia Hajj outbreak, Europe, South Africa, South America, North America, and Oceania. MenW CC ST-11 isolates from East Asia might have evolved locally. Antibiotic susceptibility tests revealed a relatively high resistance rate (22.86%) of Nm isolates to penicillin. This study provided valuable data for Chinese public health authorities to grasp the temporal epidemiological characteristics of MD and healthy carriage.


INTRODUCTION
Neisseria meningitidis (Nm), an obligate human bacterial pathogen, can cause severe, life-threatening meningococcal disease (MD) with case fatality rates (CFRs) of 10-20% even with prompt treatment (MacNeil et al., 2018), typically reported in infants younger than 1 year and adolescents/young adults (Borrow et al., 2017). Up to 20% of survivors suffer from long-term sequelae including deafness, loss of limbs, and seizure (Whittaker et al., 2017). Besides classical manifestation as meningitis or septicemia, few cases of bacteremic pneumonia, primary lumbar spondylitis, pericarditis, cellulitis, septic arthritis, and primary polyarthritis due to Nm infection have been reported worldwide, mostly in elder adults, children, or individuals with comorbidities (Yubini et al., 2018;Gomez et al., 2019).
The epidemiology of MD varies considerably across geographical locations, time periods, and age of the hosts (Pelton, 2016;Borrow et al., 2017). In recent years, most regions of the world have experienced a downward trend in the incidence of MD (Li et al., 2018). In China, the incidence has declined sharply from the peak of 403 cases per 100,000 population in 1967 to 0.047 cases per 100,000 population from 2006 to 2014 (Li et al., 2015). However, the current overall incidence rate of China might be underestimated according to subnational estimates of 1.84 cases per 100,000 population (Zhang et al., 2016).
Accompanying the decreased incidence, the prevalence of different serogroups has changed substantially globally. Control of MD caused by serogroup A has been achieved to a certain extent across the "meningitis belt" of Africa where mass immunization programs with meningococcal A conjugate vaccines (MenAfriVac) have been implemented since 2010 (Novak et al., 2019). Between 2010 and 2015, nine meningitis belt countries observed a 99% reduction in MenA disease incidence (Trotter et al., 2017). Meanwhile, MD caused by serogroups other than NmA carries a substantial disease burden in the sub-Saharan meningitis belt countries (Bozio et al., 2018;Fernandez et al., 2019). An epidemiological shift has happened from the predominance of a single serogroup to diverse causes of MD or epidemics. Successful introduction of vaccination against MenA, MenC, and, more recently, MenB, has controlled and prevented MD caused by associated serogroups effectively in European countries including the United Kingdom, the Asia Pacific region, and the African meningitis belt (Trotter et al., 2017;Li et al., 2018;Booy et al., 2019;Aye et al., 2020;Ladhani et al., 2020). However, unexpected surges in the prevalence of certain serogroups occasionally occurred, for example, MenY caused epidemics in Scandinavia between 2010 and 2012 (Broker et al., 2014), serogroup predominance switch from MenB to MenW in Chile in 2012 (Villena et al., 2019). In China, serogroup A was continuously predominant from 1960s to 1980s, responsible for more than 95% of MD cases (Hu, 1991). With the implementation of MenA polysaccharide meningococcal vaccine since 1982, the incidence rate declined steadily (Liu et al., 2000;Deng et al., 2012). On the contrary, the proportion of MenB-caused sporadic cases rose apparently (Li et al., 2015). During the course of 2015-2017, 54.08% of clinical cases were due to MenB infection (Li J. H. et al., 2019). The dynamic and unpredictable epidemiology proposed a particular challenge for MD prevention (Booy et al., 2019). Ongoing surveillance is essential for both insight into temporal MD epidemiology and to provide fundamental data for vaccine formulation, vaccine policy, and monitoring the impact of implemented vaccines.
Neisseria meningitidis is a commensal inhabitant of the human upper respiratory tract. It colonizes 2-20% of healthy individuals in the population (Chamorro et al., 2019). Healthy humans colonized by meningococci have been considered to be the main source of Nm transmission (Yazdankhah and Caugant, 2004;MacLennan et al., 2006). Under certain conditions such as the causative serogroup, the host immune system, and some environmental factors, the development of MD from carriage may occur (Kim et al., 2017). Carriage studies of Nm for healthy population will be helpful to understand in depth the circulating Nm from MD cases and carriage and detect potential hyperinvasive isolates in healthy population. It is valuable for MD control and prevention. The aims of this study were (i) to put insight into the epidemiology of both MD and carriage in Zhejiang Province, China, from 2011 to 2021 and (ii) to reveal the genotypic characteristics and antibiotic resistance of Nm isolates.

Data Collection
Surveillance of MD is mandatory in China. MD cases are notified to the National Notifiable Diseases Surveillance System by physicians nationwide since 2005. Public health specialists from local Centers for Disease Control and Prevention conduct a face-to-face investigation to collect information on demographic and epidemiological characteristics, clinical manifestation, vaccination status, laboratory test, and disease outcome for each MD case. To meet provincial monitoring requirements, investigation on the carriage rate of Nm in healthy population of all ages through throat swab has been performed in Jiaojiang, Cangnan, Yiwu, Jingdong, Longquan, Yinzhou, and Haining, respectively, during 2015-2020. Normally, subjects were divided into 7-to 8-year age groups, each of which included 30-50 individuals.

Case Definition
Confirmed case: detection of Nm nucleic acid by validated polymerase chain reaction (PCR) assay or isolation of Nm from specimen obtained from blood, cerebrospinal fluid (CSF), or skin lesion. Clinical case: present with typical symptoms without laboratory-confirmed results, such as fever, headache, vomiting, meningeal irritation, or purpura fulminans (Li et al., 2015). Associated sample collection and laboratory techniques were according to the National Surveillance Program for Meningococcal Meningitis (Ministry of Health of People's Republic of China, 2006)

Isolation, Identification, and Serotyping of Neisseria meningitidis
Blood, CSF, skin lesion, or swab samples, and so on, were directly plated onto the chocolate agar within 4 h of sample collection and incubated at 35 • C in a 5% CO 2 -enriched atmosphere for 72 h. Presumptive Neisseria colonies with typical morphology were subcultured on blood agar enriched with 5% sheep blood and subjected to further identification by using VITEK 2 compact system with NH Cards (bioMérieux, Lyon, France). If necessary, 16S rDNA was amplified and sequenced for identification (Walcher et al., 2013). Confirmed Nm isolates were serogrouped by slide agglutination using commercial antisera according to the manufacturer's instruction (Remel, United Kingdom) and PCR method (Taha, 2000;Claus et al., 2002;Bennett et al., 2004;Fraisier et al., 2009).

Genome Sequencing and Assembly
The total DNA of Nm isolates was extracted by using Bacterial DNA Kit (Omega, United States). DNA purity quotient was tested by using spectrophotometer NanoDrop TM 2000 (Thermo Fisher Scientific, Waltham, MA, United States), and integrity was assessed by using electrophoresis on 0.8% agarose gel (Thermo Fisher Scientific, United States). Purified DNA with an amount greater than 50 ng was used for sequencing library preparation. DNA was simultaneously fragmented and tagged with adapters by using the TruePrep TM DNA Library Prep Kit V2 for Illumina (Vazyme, China). Individual library was assessed on the QIAxcel Advanced Automatic Nucleic Acid Analyzer and then quantified through qualitative PCR (qPCR) by using KAPA SYBR R FAST qPCR Kits (Roche, Switzerland). The prepared library was sequenced on Illumina HiSeq X Ten platform (Illumina Inc., United States); 150-bp paired-end reads were generated. The reads were de novo assembled by using the software Unicycler. All draft genome sequences were submitted to PubMLST Neisseria database 1 .

Molecular Typing Based on Genome Sequencing
The genome sequence of each Nm isolate was submitted to PubMLST Neisseria database (see text footnote 1) and aligned to multilocus sequence typing (MLST) housekeeping genes abcZ, adk, aroE, fumC, gdh, pdhC, and pgm alleles and variants of three Finetyping antigen coding sequences variable regions porA-VR1, porA-VR2, and fetA-VR on the bacterial isolate genome sequence database (BIGSdb) platform. MLST sequence type (ST), clonal complex (CC), and Finetyping antigens subtype were determined to each isolate based on alignment results. New MLST alleles, STs, and variants of Finetyping antigens (PorA and FetA) were designated by the Neisseria profile/sequence definitions database (see text footnote 1).

Genome Comparisons
Three MenW CC ST-11 isolates (NM15, NM18, and NM19) were genome sequenced with N50 values of 63,116-75,833 and 1 https://pubmlst.org/neisseria/ contig (≥200 bp) numbers of 121-129. Genomes of additional international MenW CC ST-11 strains used for comparisons were accessible in the Neisseria PubMLST database (see text footnote 1). A total of 483 isolates from MD outbreaks, epidemics, endemics, and sporadic cases across the world were involved (Supplementary Table 1). The PubMLST genome comparator tool was used to perform genome comparisons. Analysis type was against defined loci, which include 1,605 Nm core genes in Nm cgMLST v1.0 (Bratcher et al., 2014). Minimum identity and minimum alignment were set at 70 and 50%, respectively. When the distance matrix was calculated, incomplete loci (due to incomplete assembly) were ignored in pairwise comparison, loci paralogous in all were excluded, pairwise paralogous loci were excluded, and core threshold was 90%. The phylogenetic tree based on distance matrix was visualized by using SplitsTree4 version 4.13.1 2 .

Antibiotic Susceptibility Tests
Antibiotic susceptibility of Nm isolates was assayed by using Etest strips (Antu, China). The minimum inhibitory concentrations (MICs) for each isolate to different antibiotics were measured after incubation at 35 • C for 20-24 h in 5% CO 2 on Mueller-Hinton agar with 5% sheep blood. Breakpoints for penicillin, ampicillin, azithromycin, rifampicin, minocycline, ciprofloxacin, ceftriaxone, trimethoprim-sulfamethoxazole, meropenem, levofloxacin, cefotaxime, and chloramphenicol were according to the Clinical and Laboratory Standards Institute (CLSI) documents M100 30th edition guidelines (Clinical and Laboratory Standards Institute [CLSI], 2020). Nm isolates resistant to three or more types of antibiotics belonging to different antibiotic classes were determined as multidrugresistant (Magiorakos et al., 2012). Escherichia coli ATCC 25922 was used as a quality control strain.

Statistical Analysis
Descriptive analysis was performed with SPSS 23.0. Graphs and tables were processed with Microsoft Excel 365. Correlation test for two quantitative variables was performed using the Pearson correlation test.

Epidemiology of Meningococcal Disease and Carriage in Zhejiang Province
During 2011-July 2021, 55 sporadic cases of MD were reported, 48 (87.27%) were laboratory confirmed, and 7 (12.73%) were clinically diagnosed. Overall, three cases were fatal, with an average CFR of 5.45%. The median age of those MD cases was 7 years, with the ages ranging from 12 days to 79 years old, with 31 cases aged 15 years (56.36%), occupying the highest percentage; 37 cases were male (67.27%).
During 2011-2020, the annual incidence was 0.0017-0.0183 per 100,000 population. The mortality rate was 0-0.0018 per 100,000 population. Year 2018 demonstrated the highest CFR of 25%. It was noteworthy that only one case of MD was detected in 2020, when the COVID-19 pandemic swept in China. We speculated that the strict mitigation measures against COVID-19 pandemic in China (Special Expert Group for Control of the Epidemic of Novel Coronavirus Pneumonia of the Chinese Preventive Medicine Association, 2020) might have caused a great impact on the spread and case detection of MD. For example, wearing of mask in public, limited scope of movements, and decreased willingness to move of the population could efficiently reduce the spread of respiratory infectious diseases. The highest age-specific incidence was observed in the age group of younger than 1 year (range, 0.0445-0.5013 per 100,000), followed by the age group of 1-4 years (range, 0-0.1065 per 100,000) and 10-19 years (range, 0-0.0524 per 100,000). Details are shown in Table 1.
During the study period, serogroup was identified in 30 laboratory-confirmed MD cases (62.50%  (Trotter et al., 2007;Kriz et al., 2011;Whittaker et al., 2017). The high prevalence of MenB also occurred in the United States, where more than half of MD cases among individuals aged 16-20 years were attributed to MenB (MacNeil et al., 2018). Herein, 53.3% of MD cases were due to MenB, consistent with a previous report in China (Li J. H. et al., 2019), indicating that with the extensive implementation of MenA polysaccharide vaccine (MPV-A) since the 1980s, MenA plus MenC polysaccharide vaccine (MPV-AC) since 2003, and serogroup ACWY meningococcal polysaccharide vaccine (MPV-ACWY) since 2006 (Li et al., 2018), the predominant serogroup has changed gradually from MenA and MenC to MenB in the last few years in China. Given that a vaccine against MenB is still unavailable in China, this change of serogroup prevalence might pose the potential risk of epidemics of MenB.
As to the serogroup-specific incidence, during 2011-2020 (Figure 2), MenB incidence was significantly increasing from 0.0018 per 100,000 in 2013 to 0.0070 per 100,000 in 2019 (r = 0.906, P = 0.001). MenC incidence seemed to be stable when case was detected. MenY cases occurred in 2019 with the incidence rate of 0.0017 per 100,000.
Although Zhejiang Province had carried out the surveillance on carriage rate of Nm in healthy population since 2010, positive pathogen from throat cultures was first isolated in 2015. During 2015-2020, 17 positive culture isolates were detected from 2,827 throat swabs (0.64%) ( Table 2). Analysis revealed

Characteristic
Cases  that the distribution of serogroups showed high homogeneity in healthy population, 12 samples (70.59%) were MenB, and 1 sample (5.88%) each for MenC and MenY. MenB was the most predominant serogroup both in MD cases and in healthy population. In addition, three carriage isolates were identified as untypable, with a percentage of 17.65%. According to previously published reports, we found that the proportion of untypable Nm isolates from carriages was variable in different geographic regions and different populations. For example, the percentage of untypable Nm isolates from the 1-to 29-year-old population in Burkina Faso, one of the African meningitis belt countries, before the introduction of a meningococcal serogroup A conjugate vaccine was 13.3% (Kristiansen et al., 2011), which was close to that in our study. Meanwhile, in Paraguay, the proportion attained 31% in children and young adults (Chamorro et al., 2019). The majority of Nm isolates from healthy population in seven countries of the African meningitis belt (52%) (MenAfriCar Consortium, 2015) and Southern Ethiopia (76.4%) were untypable (Bårnes et al., 2016). So the proportion of untypable Nm isolates from carriage might be relevant to geographic regions and population.
The median age of subjects with MenB was 23 years, and the range was from 9 months of age to 50 years old. Approximately 42% of subjects with MenB were aged 30-59 years, and 33% of

Multilocus Sequence Typing and Finetyping Antigen Analysis
Thirty-five Nm isolates from both MD cases and healthy population were further molecular typed (  (Zhou et al., 2012). MenB CC4821 was also dominant among serogroup B strains. Different from MenC CC ST-4821, MenB CC4821 was usually associated with sporadic infections (Zhou et al., 2012). A MenX ST7 (CC ST-5) strain was isolated from an MD case in Wenzhou in 2013. Our previous studies (Pan et al., 2014;Ji et al., 2017) have determined that this strain arose from capsule switching between a MenX and a MenA ST-7 strain. Nm is naturally competent for exogenous DNA uptake and horizontal exchange in all growth phases, which enables gene exchange at the capsule polysaccharide biosynthesis locus between different strains resulting in capsule switching. Four MenW isolates were detected from MD cases. Three were grouped into CC ST-11, and one belonged to CC ST-4821. The hyperinvasive MenW CC ST-11-caused cases, epidemics, or even pandemics have emerged around the world. In 2000, a Hajj-linked MenW CC ST-11 outbreak swept the Middle East, the meningitis belt of Sub-Saharan Africa and South Africa (Mustapha et al., 2016). In recent years, endemic MenW CC ST-11 disease increased in South America, North America, Europe, and China (Zhou et al., 2013;Tsang et al., 2017;Eriksson et al., 2018;Rubilar et al., 2018). It was worth noting that two MenY CC ST-23 sporadic cases occurred in Taizhou and Haining in 2019 and 2021, respectively. All associated isolates belonged to ST1655 (Table 1). It has been known that most of the MenY MD was attributed to CC ST-23 (Hedberg et al., 2011;Toros et al., 2014). In Sweden, 62% of serogroup Y isolates from MD cases were CC ST-23 (Toros et al., 2014). During 1989-1991, approximately 2% of MD strains belonged to serogroup Y in the United States, whereas, by the mid-1990s, the percentage rose up to more than 30%. The increased proportion of MenY among all MD was attributed to CC ST-23 (Jackson and Wenger, 1993;Rosenstein et al., 1999;Harrison et al., 2006). Until early 2019, there were no MenY CC ST-23 isolates reported in China. MenY isolates in China were mostly associated with healthy carriers and grouped into three CCs, CC ST-175, CC ST-92, and CC ST-198 (Zhu et al., 2015;Wang et al., 2017;Shan et al., 2018). In March 2019, MenY CC ST-23-caused MD case first emerged in Guangdong Province, China (Li B. et al., 2019). MenY CC ST-23 MD cases increased gradually in China.

Genome Sequencing-Based Phylogenetic Analysis of MenW CC ST-11 Strains
Since a Mecca Saudi Arabia Hajj-associated outbreak in 2000, Hyperinvasive MenW CC ST-11 has swept the globe (Ladhani et al., 2015;Mustapha et al., 2016). In China, before 2011, only three cases of MenW MD were reported (Shao et al., 2010), whereas from February 2011 to June 2012, MD caused by MenW increased (11 cases total). All associated isolates were identified as CC ST-11 (Zhou et al., 2013). Herein, based on genome sequencing, the phylogenetic relationship of MenW CC ST-11 isolates from China with international FIGURE 3 | Core genome multilocus sequence typing (MLST) neighbor-net phylogenetic tree for the distribution of MenW CC ST-11 isolates from China and additional international isolates. The tree was constructed with SplitsTree4 version 4.13.1 (http://www.splitstree.org). and 32 isolates (91.43%), respectively. In the empirical MD therapy scheme, the third-generation cephalosporins (e.g., ceftriaxone and cefotaxime) were recommended as the treatment of choice. Our results and previous reports (Gorla et al., 2018a) showed that Nm isolates with different sources had different cephalosporin susceptibility frequency, such as ranging from 100 to 64.2% for cefotaxime. Penicillin also remains one choice for MD treatment, although the increased resistance of Nm isolates to this antibiotic from different countries has been reported (Harcourt et al., 2015). To date, the resistance profiles to penicillin of Nm isolates in China were still ambiguous. According to the results herein, a relatively high level of penicillin resistance was observed. Fluoroquinolones, mainly including ciprofloxacin, levofloxacin, and norfloxacin, have been used for MD prophylaxis for high-risk adult close contacts to prevent secondary MD cases and outbreaks (Chen et al., 2015;Gorla et al., 2018b). Compared with previously reported resistance rates (0.0-84%) to ciprofloxacin of Nm isolated from different regions of different countries (Chen et al., 2015;Gorla et al., 2018b;Tefera et al., 2020), a consistent resistance rate was observed among our isolates to that from Shanghai, China, since 2005. Nm isolates from China presented a significantly higher ciprofloxacin resistance level than that from Brazil, South Korea, and so on. Among the ciprofloxacin-resistant isolates, 62.50% (20/32) were assigned either CC ST-4821 (n = 17) or CC ST-11 (n = 3), including eight (22.86%) multidrugresistant strains, resistant to ciprofloxacin/penicillin/ trimethoprim-sulfamethoxazole/levofloxacin and rifampin/ ciprofloxacin/penicillin/trimethoprim-sulfamethoxazole/levofloxacin, respectively.

CONCLUSION
In summary, a comprehensive study of epidemiology of MD, healthy carriage, and characteristics of Nm isolates in Zhejiang Province, China, was performed, providing valuable data for understanding the epidemiological characteristics of MD and carriage, as well as the need for and impact of vaccination. During the study period, the annual incidence of MD remained relatively low, ranging from 0.0017 to 0.0183 per 100,000 population. The highest age-specific incidence was in the group younger than 1 year. MenB presented predominantly in both MD cases and healthy population; a significant increase in MenB incidence was also observed from 2013 to 2019. Combined with previously published data, the dominant proportion and increasing incidence of Men B should cause more concern in China. It was worth noting that two MenY CC ST-23 cases occurred in 2019 and 2021. MenY CC ST-23 MD cases increased gradually in China. To the best of our knowledge, this is the first time that the phylogenetic relationship of MenW CC ST-11 isolates from China with international isolates was analyzed to speculate the origin of Chinese isolates based on genome sequencing. In addition, the relatively high resistance rate (22.86%) to penicillin among Nm isolates and the discovery of multidrug-resistant strains indicate a potential public health problem.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://pubmlst.org/ bigsdb?db=pubmlst_neisseria_isolates, 105802-105807.

AUTHOR CONTRIBUTIONS
YZ and XD: experimental operation, data processing, and manuscript writing. YJ: genome sequencing and data processing. JZ, LZ, LM, and HL: experimental operation. PY and HH: manuscript writing and revision. All authors contributed to the article and approved the submitted version.