In total, 4 strains have the MICs against erythromycin lower than 0.023 ug/ml, which refers to sensitive to erythromycin in vitro. The rest of the 200 strains were all resistant to erythromycin with the MICs≥256 ug/ml. All the resistant strains posed an A2047G mutation in 23S rRNA and no mutation occurred this site of the sensitive strains.
Among the 702 NPs for sequencing, 480 obtained both the available sequencing results of 23 rRNA and ptxP . All the sequencing results from the strains were as same as from the related NPs. Combined with the results from strains, there were 449 in 480 specimens (93.5%) shown the allele G in 2047 site of 23 rRNA that defined as erythromycin resistant B. pertussis infection, which also shown the allele of ptxP1. The dynamic changes of proportions of circulating B. pertussis from 2012 to 2016 as shown in figure 1.
Furthermore, 47 patients were excluded when analysis the difference among demographic characteristics cause of unclear vaccination status. Only 2 of the 21 ptxP3 strains infected in children vaccinated with co-purified aPV, that showed a significant difference between the ptxP1 strains does (c2=6.87, P=0.032). All the vaccinated subjects were administrated with co-purifid aPV (Table 1).
Within our study, we discovered that ptxP1-ER strains have been steadily increased to the circulating B. pertussis population from 2012 to 2016 in Xi’an, China. Moreover, unlike what happened to purified aPV has been administrated that B. pertussis could not only infect the infants that were too young to be vaccinated, but also the infants vaccinated with the purified aPV [11, 12], the ptxP3 strains rarely infected the infants administrated with co-purified aPV from our study.
The increasing incidence of pertussis was also emerged in China from 2013 according to the national infectious diseases case reported system. Besides the A2047G mutation in 23S rRNA occurred in ptxP1-ER B. pertussis strains, a novel fhaB C5330T was also founded in all these strains. This fhaB3 lineage has been proved to be prevalence among China via expansions most likely due to antibiotic pressure[3]. This study also illustrated that the ptxP1/fhaB3-ER strains might be adapted to the co-purified aPV. Whether the fhaB C5330T contributed to this adaption need to be further investigated.
According to this study, ptxP3 strains with the decreased proportions have observed from 2012 to 2016 in Xi’an, the western of China. In China, the co-purified aPV was free and predominated used since 2006 while the purified aPV (Sanofi) was available by paid since 2011with rarely market supplied, especially in undeveloped regions of western China, such as Xi’an. The rare of ptxP3 strains in Xi’an after 10 years of co-purified aPV used also indicated that the co-purified aPV did not give the adaption as purified aPV did in developed countries where ptxP3 was quickly predominant worldwide [13].
The co-purified aPVs have more protein antigen than purified aPVs [14]. Therefore, this study further supported the hypothesis that the small antigen targets of purified aPV could induce the vaccine pressure and vaccine adaption more easily than the more antigen targets vaccine, such as wPV, even the co-purified aPV [15]. Furthermore, among the additional protein antigens of co-purified aPV, most was the out membrane proteins such as BipA and SphB1. Such membrane proteins containing in the out membrane vesicle (OMV) of B. pertussis have been suggested as an attracting candidate component of the possible new modified vaccine against pertussis [16, 17]. The latest study further proved that the OMVs can protect against B. pertussis with long term duration, even the global popular ptxP3 and pertactin deficient strains [17].
Japan was the first country to develop aPV (co-purified) in 1981 and to adopt for use in the general population. It has been reported that both of the two types of aPVs was used recently [18]. However, the ptxP3 lineage still holds lower than 50% from 2006 to 2010 until the period of 2011-2014which reached close to 80% [19].
Most of the cases of this study were from the west of China. Otherwise, it was reported that the ptxP1-ER strains contributed to 75.4%, 50.7% and 48.6% in the circulating strains in Zhejiang province (Southern of China, 2016), Shanghai (Southern of China, 2016-2017) and Shenzhen (Southern of China,2015-2017), while the rest ES strains were almost ptxP3 strains [4, 20, 21]. No details of the vaccine type were described in these relative high proportion of ptxP3 areas of China. Liking what happened to Japan, we assumed that the purified aPV used was much more in these developed areas of China than in Xi’an, which generate a relative low level of vaccine protection from co-purified aPV in general population. As a result, the proportions of ptxP3 strains were much more.
Consistent with reports in these areas of China, the erythromycin resistant strains were almost ptxp1 allele while the ptxP3 strains were all sensitive to erythromycin. As shown in this study, though the average age of ptxP3- ES strains infection group is lower than in ptxP1- ER groups, there is no significant difference. Furthermore, more than 85% of subjects have taken antibiotics before sampling and detection, no difference was observed between the ptxP3- ES and ptxP1- ER groups (data not shown). Therefore, despite the antibiotic pressure which seems to provide the selective advantage for expansion of erythromycin resistant strains, we suggested that the co-purified aPV protect against ptxP3 strains more efficient, which generated a rare chance for ptxP3 strains to be under the antibiotic pressure and further developed to be erythromycin resistance.
However, it is a limitation that the cases of ptxP3 strains were relative too small to give strong evidence about the protection against ptxP3 lineage by co-purified aPV. Furthermore, the ptxP3 with the pertactin (PRN) deficient isolates were widely appeared in some industries countries [22], if the ptxP3 isolations in this study expressed of PRN were unknown in this study. Lastly, the age of the patients in our study was mainly the infant but not many children after at least 5 years of vaccination of co-purified aPV. Thus we can not give powerful support about the protection duration against ptxP3 linage of the co-purified aPV.