The characteristics of current natural foci of hemorrhagic fever with renal syndrome in Shandong Province, China, 2012-2015

Background Hemorrhagic fever with renal syndrome (HFRS), an infectious disease caused by hantaviruses, is endemic in China and remains a serious public health problem. Historically, Shandong Province has had the largest HFRS burden in China. However, we do not have a comprehensive and clear understanding of the current epidemic foci of HFRS in Shandong Province. Methodology/principal findings The incidence and mortality rates were calculated, and a phylogenetic analysis was performed after laboratory testing of the virus in rodents. Spatial epidemiology analysis was applied to investigate the epidemic foci, including their sources. A total of 6,206 HFRS cases and 59 related deaths were reported in Shandong Province. The virus carriage rates of the rodents Rattus norvegicus, Apodemus agrarius and Mus musculus were 10.24%, 6.31% and 0.27%, respectively. The phylogenetic analysis indicated that two novel viruses obtained from R. norvegicus in Anqiu City and Qingzhou City were dissimilar to the other strains, but closely related to strains previously isolated in northeastern China. Three epidemic foci were defined, two of which were derived from the Jining and Linyi epidemic foci, respectively, while the other was the residue of the Jining epidemic focus. Conclusions/significance The southeastern and central Shandong Province are current key HFRS epidemic foci dominated by A. agrarius and R. norvegicus, respectively. Our study could help local departments to strengthen prevention and control measures in key areas to reduce the hazards of HFRS.

According to previous studies, over 90% of all cases of HFRS in the world have occurred in China in recent decades [4,5]. Although the harm caused by HFRS in humans has declined in recent years due to the development of preventive measures and medical capabilities, HFRS is still a serious public health threat in China today. Shandong Province, which has the second largest population among all the provinces in China, has been suffering from a large HFRS burden for several decades. In the past decade, Shandong Province remained in the top five of all provinces in China regarding the number of HFRS cases [1].
Changes in the epidemic foci (areas in which some wild animals have a long-term preservation of infectious pathogen) of infectious diseases are caused by many factors, including natural, human and viral host factors [6]. In recent years, Shandong Province developed an R. norvegicus-dominated mixed viral genotypes epidemic focus. However, the characteristics of the HFRS epidemic foci in Shandong Province have been constantly changing, and the health threat to humans has been constantly changing as well. Adjustments to policies and practices should be performed in a timely manner. However, although many studies have analyzed the HFRS epidemic situation in Shandong Province [4,6,7], we do not have a comprehensive and clear understanding of the current epidemic foci of HFRS. For this reason, we conducted a systematic analysis of surveillance information on HFRS in Shandong Province from 2012 to 2015 to assess the severity of the situation and determine the new epidemic foci in Shandong Province.
work described here adhered to the guidelines of the Animal Subjects Research Review Boards at the Beijing Institute of Microbiology and Epidemiology.

Study area and data source
The study area is located in the North China Plain, in eastern China (34˚22'to 38˚23'north latitude, 114˚19' to 122˚43' east longitude, Fig 1), with a population of more than 99 million and an area of 158,000 km 2 . Shandong Province has a large population of agricultural workers and high land cultivation rate in China. The central region is mountainous and hilly, which is surrounded by plains.
The records of HFRS cases between 2012 to 2015 were obtained from the China Information System for Disease Control and Prevention and all data analyzed were anonymized. All patients were diagnosed as HFRS clinically. established with consideration of the severity of the local HFRS epidemic and geographic landscape differences. Rodents surveillance data were also collected from the China Information System for Disease Control and Prevention. Lung samples of the captured rodents were obtained for laboratory testing.

Epidemiological description
The 2012-2015 records of HFRS cases covering the whole of Shandong Province were used to calculate the monthly incidence rate (per 100,000 population). The proportions of cases in different demographic groups, based on gender, age and occupation were calculated. To observe the temporal distribution of cases more intuitively, we divided all cases into three groups (Feb to May, June to Sept and Oct to Jan) based on the two annual HFRS incidence peaks.

Rodent surveillance analysis
The night trapping method was used to capture rodents with peanuts as bait. The wild and residential areas were both taken into account when placing the rat clips and spring and autumn were selected as the sampling time for each surveillance site. The sampling process of each surveillance site was carried out by professionals use same method to ensure the representativeness of results. According to the information on the captured rodents, the proportion of each rodent species out of all the rodents captured was calculated. Lung samples of the captured rodents were screened using the reverse transcription-polymerase chain reaction (RT-PCR). Total RNA, which was extracted from 20-50 mg of lung tissue using the TRIzol reagent (Invitrogen, USA), was reverse-transcribed using M-MLV reverse transcriptase (Invitrogen, USA) and the primer P14: 5'-TAGTAGTAGACTCC-3' [8]. The PCR was performed to amplify a partial L sequence, using primer pairs as previously described [9]. Amplification result was sequenced and compared with data from the National Center for Biotechnology Information (NCBI) database using the Basic Local Alignment Search Tool (BLAST) to determine which genotypes they were [10].

Phylogenetic analysis
It was found that the strains of same surveillance sites stacked in same small branch based on the results of preliminary phylogenetic analysis. Therefore, we screened the strains with low homology in the same county every year. Eventually, the partial L segments of 29 strains from the 14 surveillance sites were selected (S1 Table and S1 File). Additionally, we searched Gen-Bank and downloaded data on 18 relevant HV strains isolated in other regions of China, the Korean Peninsula and Europe. A phylogenetic tree was then constructed, using the Maximum likelihood (ML) method and bootstrap testing (1000 replicates), in MAGE 7.0 software (Pennsylvania State University, PA, USA) [11].

Epidemic foci analysis
According to the HFRS cases distribution, molecular epidemiology and rodent surveillance data, the epidemic foci of HFRS in Shandong Province during the study period were defined. The Kriging interpolation method was applied to display the extent of the epidemic foci more clearly.

Data analysis
Data were analyzed using SAS 9.4 (SAS Institute Inc., Cary, NC, USA). All maps were mapped by using a geographical information system (GIS) technique in ArcGIS 10.2 software (ESRI Inc., Redlands, CA, USA). MEGA 7.0 was used to conduct the phylogenetic analysis.

Epidemiological characteristics of HFRS
From 2012 to 2015, 6,206 HFRS cases and 59 related deaths were reported in Shandong Province, and the annual incidence rate, mortality rate and case fatality rate ranged from 1.15 to 1.87 per 100,000, 0.01 to 0.02 per 100,000, and 0.80% to 1.07%, respectively (Table 1). During the study period, the annual HFRS incidence rate in Shandong Province peaked in 2013, then decreased year by year, and reached its lowest point in 2015. In addition, there were two incidence peaks each year, the small peak was from April to June and the larger one was from October to January. However, mortality rates and case fatality rates declined year by year. The deaths mainly occurred from September to January (Fig 2). As for spatial distribution, significant dynamic variation was found in that the high-risk areas were concentrated in a single large region in central and southeastern Shandong Province in 2012, which gradually separated into two regions, and then eventually formed two relatively independent high-risk regions in 2015 (Fig 3).
Regarding the demographic characteristics, 4465 males accounted for 72% of all cases. We found that 81% of all cases occurred in individuals aged 30-70 years, with the 41-50-year-old age group having the highest proportion of all the 10-year age groups. The three occupations with the highest incidence rates were farm workers (85%), urban workers (5%) and students (3%). The results clearly showed that middle-aged male farm workers are the main population to experience HFRS in Shandong Province. ( Table 2). The laboratory testing results revealed that 111 of the captured rodents were positive for HV, of which 93 were R. norvegicus, 17 were A. agrarius and 1 was Mus musculus. The mean HV carriage rate among all rodents was 6.17%, and the rates of R. norvegicus, A. agrarius and Mus musculus were 10.24%, 6.32% and 0.27%, respectively (Table 3). It can be seen that the HV carriage rates in this study were highest for R. norvegicus and A. agrarius among all the rodent species. However, it is worth noting that one HV strain was also isolated from a single Mus musculus in this study. We constructed a geographic map of the different HV-infected rodent species. This indicated that the HV-infected dominant rodents in central Shandong Province were R. norvegicus, followed by Mus musculus, while A. agrarius was the dominant HV-infected rodent in southeastern Shandong Province ( Fig 4A). Additionally, the HV carriage rates of rodents in each surveillance site were calculated and plotted on another map ( Fig 4B). This showed that the surveillance sites with high HV carriage rates were mainly concentrated in central Shandong Province (Guangrao county had the highest HV carriage rate, at 16.67%), while the surveillance sites in southeastern Shandong Province had lower HV carriage rates (Pingdu city had the highest HV carriage rate, at 6.67%). Obviously, there were distinct differences in the spatial distributions of different HV-infected rodent species and HV carriage rates in Shandong Province.

Phylogenetic analysis
A total of 111 samples generated the PCR products of expected size for the partial L sequence. from the rodents captured in Shandong Province from 2012 to 2015. The Blast results using GenBank revealed that the numbers of SEOV and HTNV strains were 91 and 18, respectively, and two strains with novel genotypes were also found.
To construct a phylogenetic tree, partial L segment sequences from 47 HV strains were used: 29 strains from our samples and 18 strains from GenBank (which had been isolated in mainland China, the Korean Peninsula and Europe). As shown in Fig 5, the branches of the phylogenetic tree revealed that the HTNV strains obtained in this study consisted of two clades from a bigger branch. They shared a common ancestor with strains isolated in Hubei and Xi'an. The SEOV strains obtained in this study constituted one branch, except for one strain that was obtained in Jiaxiang County (in the southwest of Shandong Province). Almost all the SEOV strains were closely related to viruses isolated in Fujian, Beijing and Zhejiang. However, the SEOV strain obtained in Jiaxiang County was dissimilar to all other strains and formed a separate branch. In addition, the genetic diversity associated with the SEOV branches indicated that they were more homogeneous than the HTNV branches, which had a higher degree of genetic diversity.
We obtained two viruses with novel genotypes from R. norvegicus in Anqiu City and Qingzhou City that were different from the other strains. After the alignment analysis in GenBank, it was found that they were closely related to the strain previously isolated in Russia (Jewish Autonomous Oblast), which has not been classified and named. Then together with strains isolated in Fusong County and Shenyang City (northeastern China), forming a single branch in the phylogenetic tree. Previous studies on the complete S segments of strains isolated in Fusong County and Shenyang City have shown that they are closely related to a virus isolated in Vladivostok in the Russian Far East, forming a single branch. The researchers believed that they belong to the same HV genotype, the Vladivostok virus (VLAV), which host are Microtus fortis [12]. Therefore, our two novel viruses appear to be similar to the VLAV genotype [12,13,14]. We also obtained HTNV (1584101 HD 2015) from the wild-collected R. norvegicus in Huangdao. This result suggested that HTNV in A. agrarius has spilled over to R. norvegicus, which was previously not common in Shandong Province.

Analysis of HFRS epidemic foci
To more intuitively identify the current HFRS epidemic foci in Shandong Province, the distribution of HV genotypes in each surveillance site was mapped. Additionally, the time distribution of HFRS throughout Shandong Province has a seasonal periodicity obviously. Peaks generally occur in spring or autumn and winter. Based on the two annual incidence peaks, we divided the 2012-2015 cases into three groups according to the month of their occurrence (Feb to May, June to Sept, and Oct to Jan) and mapped them with the 2012-2015 annual incidence rates on a single map. Finally, as shown in Fig 6, we summarized the 2012-2015 HFRS case spatial epidemiology, molecular epidemiology and rodent surveillance results in Shandong Province, indicating the epidemic foci during the study period.
The epidemic foci comprised two major foci, designated Regions a and b, which is located in southeastern and central Shandong Province, respectively, and a smaller one designated Region c in southwestern Shandong Province (S2 Table). Among them, Region a had the largest coverage area and the highest incidence rate, with most cases occurring in the autumn and winter, and the rodents carrying the virus were mainly A. agrarius. The viral genotypes in Region a were mainly HTNV, which constituted an A. agrarius-dominated mixed-type epidemic focus. The coverage and incidence rates of Region b were smaller than in Region a, the Epidemic foci of hemorrhagic fever with renal syndrome in Shandong Province temporal distribution of cases in Region b was relatively uniform, and the infected rodents were mainly R. norvegicus. The viral genotypes in Region b were mainly SEOV, which constituted an R. norvegicus-dominated mixed-type epidemic focus. The coverage and incidence rates of Region c were much smaller than in Regions a and b. The cases in this area occurred mostly in late spring and early summer, and the species of infected rodents and viral genotypes were more similar to those of Region b, involving an R. norvegicus-dominated mixed-type epidemic area.   Epidemic foci of hemorrhagic fever with renal syndrome in Shandong Province

Discussion
As a natural epidemic disease, HFRS has a prominent position in the history of infectious disease prevention and control in Shandong Province. In Shandong Province, HFRS case peaked in the mid-1980s, then declined gradually and then became stable [6,15]. Based on the 2012- Epidemic foci of hemorrhagic fever with renal syndrome in Shandong Province 2015 HFRS data from Shandong Province, the characteristics of the disease, virus host and laboratory testing were analyzed in this study, and the nature and characteristics of current epidemic foci were identified. Overall, the annual HFRS incidence rate exhibited small fluctuations during the study period, peaking in 2013, then decreasing year by year, and reaching its lowest point in 2015. There were still annual incidence peaks, which comprised a small peak in late spring to early summer and a big peak in autumn to winter. The small peak is associated with indoor-related infections linked to R. norvegicus breeding activity, while the big peak is mainly due to the increased contact between people and wild rats when they work in fields, and the migration of wild rats from fields to residential areas for foraging in the later period [15].
It is worth noting that 88% of all deaths occurred in September to January. Previous studies have suggested that the clinical symptoms of HFRS caused by HTNV are often more serious than those caused by SEOV [16]. Additionally, human immunity decreases in autumn and winter when it is cold. Therefore, the deaths may be the result of a combination of HTNV and seasonal factors. Additionally, male farmers represented the biggest HV-susceptible population because of their increased contact with rodents due to working in the fields and poor residential sanitation conditions [17]. Therefore, HFRS prevention and control in Shandong Province should continue to be focused on controlling the rodents' population and strengthening the immunization and education interventions for farmers in key epidemic foci. To reduce the mortality rate, we should also pay more attention to the treatment of cases occurring in autumn and winter.
We found that although the HV carriage rate among rodents was lower in southeastern Shandong Province than in central Shandong Province, the epidemic situation and clinical symptoms of cases were worse. This suggests that more attention should be paid and more interventions should be undertaken in the southeastern region. Additionally, the proportion of Mus musculus carrying HV is very low in this study and other researches. Therefore, determining whether Mus musculus can carry HV normally and pass on the infection to humans should be a key task of future research. [18,19]. Epidemic foci of hemorrhagic fever with renal syndrome in Shandong Province HTNV and SEOV were still the main HV genotypes in Shandong Province. Although we inferred the patient's situation according to the HV genotype from rodents, studies have shown that HV isolated from patient's serum in Shandong Province was associated with local dominated rodent species. For example, the serum test results of Zhaoyuan patients showed that the area belongs to Apodemus-type epidemic focus, while Jining, Zibo, Weifang belong to Rattus-type epidemic foci. These conclusions are consistent with this study [20,21]. But two points are worth noting. First, we obtained HTNV from R. norvegicus captured in the wild, indicating that HTNV from A. agrarius spilled over to R. norvegicus. Previous studies also showed that some HV obtained from R. norvegicus seemed to be HTNV [13,22,23]. However, each HV genotype is associated primarily with a specific rodent host species that it coevolved with [18,24,25]. The phenomenon of viruses spilling over to non-traditional specific rodent hosts can allow them to find more suitable surroundings for a higher survival rate and more efficient infection and replication [26,27]. Huangdao is one of the most serious HFRS epidemic foci, which is located in southeastern Shandong. This phenomenon may promote the epidemic situation in this region. It is perhaps one of the reasons for migration of geographic boundaries of HFRS endemic areas. Second, we found two novel strains obtained from R. norvegicus in Anqiu City and Qingzhou City and they were closely related to strains from northeastern China, which reportedly belong to the VLAV genotype. This phenomenon of potential VLAV strains in Shandong Province has not previously been reported. Therefore, determining whether the viruses were transmitted from northeastern China, or simply involve indigenous genetic mutations, requires further research.
In the 1970s, HFRS was in completely Apodemus-type natural focal state in Shandong. Then the Apodemus-type focus in the southeastern part of Linyi and the Rattus-type focus in the southern part of Jining began to expand and merge with each other, and formed two foci in the 1980s, one mixed focus dominated by Apodemus-type in the hilly area of the southern and middle part of Shandong, another one dominated by the Rattus-type in the Yellow River valley of the northwestern part of Shandong. In the 1990s, Shandong Province formed a mixed focus dominated by Rattus-type, and entered a stable stage [15].
The results of Liu and Fang showed that the epidemic focus of HFRS in Shandong Province before this study was mainly concentrated in the northern part of Linyi in southeastern Shandong Province with a dominance of spring peak of incidence [6,7]. The whole landscape of Shandong Province became a uniform mixed HTN-type and SEO-type endemic focus with dominant rodents R. norvegicus and Mus musculus. However, the current epidemic foci of HFRS in Shandong Province shows changes both in its distribution and the composition of the HV host. In terms of the temporal distribution, the influence of Region a, which involved an A. agrarius-dominated mixed-type epidemic focus, on HFRS epidemic situation in Shandong Province mainly occurred in the autumn and winter. The coverage of Region a was more extensive than the coverage of Regions b and c, and the epidemic situation was also more serious. We inferred from previous studies that Region a is derived from the epidemic foci previously centered on Linyi City. Conversely, Regions b and c were more similar, involving R. norvegicus-dominated mixed-type epidemic foci. In Regions b and c, the temporal distribution of HFRS cases was more uniform and the epidemic situation was more moderate than in Region a. It was inferred that Region b is derived from the epidemic foci previously centered on Jining City, while Region c is the residue of the epidemic foci centered on Jining City [7].
Although the HV carriage rates among rodents in Regions b and c were higher than that associated with Region a, the severity of the epidemic situation was less severe, suggesting that the HV-susceptible population appears to be more susceptible to the HV strains in Region a (which mainly belong to the HTNV genotype). Interestingly, the distance between Regions b and a is much smaller than that between Regions b and c. Despite this, the epidemic characteristics of Region b are quite different from those of Region a, and more similar to those of Region c. Although previous studies have shown that HTNV (which was the main genotype in Region a, where the epidemic situation was more serious) is more harmful to humans than SEOV (which was the main genotype in Regions b and c), local geographic conditions, climate, HFRS prevention and control measures and safety awareness among susceptible populations are also important factors [16,[28][29][30][31][32]. Therefore, we believe that a further study should be performed to explore the factors underlying our result.
Several limitations of this study should be noted. First, the HV genome can be divided into S, M and L segments. A comprehensive phylogenetic analysis of the three segments would give us a clearer understanding of the genetic diversity and evolutionary trajectory of HV. The data on S and M segments in this study were insufficient, so the phylogenetic analysis of HV may be biased. Second, there are many factors that can influence the distribution and severity of infectious disease, such as natural environmental factors, urbanization processes and local prevention efforts [32][33][34]. However, this research lacked information on these factors, so the HFRS spatio-temporal distribution analysis is inadequate.
In conclusion, the southeastern and central Shandong Province are current key HFRS epidemic foci dominated by A. agrarius and R. norvegicus, respectively. Although the overall epidemic intensity was lower than the previous intensity in Shandong Province, it was still relatively high compared to the rest of the country. The spatio-temporal distribution of HFRS within Shandong Province indicated different epidemic situations in different regions. Local departments still need to strengthen their prevention and control measures to reduce the hazards of HFRS.
Supporting information S1 Checklist. STROBE checklist. (DOC) S1 File. The partial L segment sequence of the strains without the accession number in this study.