G群β-溶血鏈球菌之快速菌種鑑定、分子流行病學和抗藥基因分析

Abstract

β溶血的group G鏈球菌主要包括了兩種菌種：Streptococcus anginosus，Streptococcus dysgalactiae subsp. equisimilis，是正常人皮膚、咽喉、腸道等的常在菌，主要會造成免疫力不良的病人之菌血症、心內膜炎、咽喉炎、鏈球菌休克症候群（streptococcal toxic shock syndrome）等等。臨床上，對於G群鏈球菌只有做到Lancefield的分群，很少有做到菌種的鑑定。但是由於兩種菌的臨床意義不盡相同，因此菌種的鑑定不容忽視。本實驗利用anginosus group特異性的引子，對1998年~2004年台大醫院所收集到的408株G群鏈球菌進行PCR，配合傳統的鑑定方式，可以快速又正確的鑑定出G群鏈球菌中的兩種菌。實驗結果發現，S. dysgalactiae subsp. equisimilis佔了收集菌株的68%，數目是S. anginosus 31%的兩倍，且從檢體來源可以看出較S. anginosus更具侵入性。由於許多報導指出S. dysgalactiae subsp. equisimilis的emm分型與細菌的侵入能力有關，同時從菌株資料的分析中發現有14位病人有重複感染G群鏈球菌的情形，因此希望藉由對emm和脈衝式電泳的分型，進一步探討特定分型與疾病的相關性。從結果可知，幾乎所有的S. dysgalactiae subsp. equisimilis（除了1株例外）都可以做emm分型，且以stG6.1和stG485的11.27 %為最多數（n=46），S. anginosus則只有0.85%（n=13）的菌株可以分型；此外，造成重複感染的病人以stG485和stG6.1的S. dysgalactiae subsp. equisimilis占最多，由此可見帶有特殊emm的基因可以幫助細菌對宿主的感染和抵抗宿主的免疫反應。為了了解G群鏈球菌對於藥物的感受性，及調控其抗藥的基因為何，又進行了18種抗生素的藥物感受性試驗，並針對抗藥菌株做抗藥基因的分析。結果發現，只有約20%的菌對azithromycin和clindamycin有抗性，而22%的菌對quinupristin-dalfopristin有抗性，其他的藥物MIC範圍都很低，表示臨床第一線用藥對G群鏈球菌仍然是有效的。此外，從double disc的試驗發現到50株azithromycin抗藥、clindamycin有敏感性的菌株中，有2株（4%）菌是誘發型的iMLSB，其餘的48株（96%）是M表現型；另外，加上14株azithromycin和clindamycin都抗藥的菌株之抗藥基因分析，發現7.8%的菌株具有erm(B)的基因，4.7%的菌株有erm(TR)的基因，75%的菌株有mef(A)的基因，而有3.1%的菌株同時有erm(B)和mef(A)的基因。最後，由於在S. bovis的erm(T)抗藥基因上下游有IS1216的存在，因此想知道G群鏈球菌中是否有類似構造。從長片段PCR和TA cloning的選殖、定序，知道IS1216內所帶的是一個tet(S)抗藥基因，其結構與Lactococcus lactis subsp. lactis pk214所帶的IS1216-tet(S)相似；而且經由PCR篩選，找到共有12株菌帶有這種類似的構造。利用不同限制酶的南方墨點法，得知此結構應為single copy；另外，從Apa I的脈衝式電泳對12株菌分型，可以知道有2組菌（分別有7株和2株菌）是同一種分型；使用I-Ceu I跑脈衝式電泳，轉印到膜上做南方墨點法，可以確認此段IS是位在細菌的染色體上。利用LA PCR定出其中三株菌之tet(S)上下游序列，再使用PCR預測其他菌株的結構，可以知道有9株菌上游的IS1216有缺損，但下游的則是完整的，更上游有一個缺損gamma基因，更下游有一個缺損的aadE和putative的蛋白；有2株菌的上下游IS1216都缺損，可能是屬於同一個clone；最後1株菌的上下游IS1216都完整，更上游有兩個不知功能的基因，更下游則是與Lactococcus lactis subsp. lactis pk214的IS1216下游一致。接合生殖的結果並沒有發現這樣IS1216-tet(S)的結構可以在菌種間傳遞，但可以推論，這些菌株可能遇到同一種帶有原始IS1216-tet(S)的跳躍子或質體，在嵌入染色體的過程中，發生了不同程度的缺損。

Human isolate β-haemolytic group G streptococci (GGS) include two major species- Streptococcus anginosus, Streptococcus dysgalactiae subsp. equisimilis. These bacteria are usually regarded as normal flora colonized human skin, throat, intestine, and may cause serious disease such as bacteremia, endocarditis, pharyngitis, streptococcal toxic shock syndrome on immunodeficiency patients. Clinical identification of group G streptococci with conventional methods is usually to the group level, not the species level. However, species identification should not been ignored because of the different clinical significance between two species. In this report, we collected 408 clinical isolates from 1998 to 2004 in NTU hospital and used anginosus specific primers combined with conventional methods to rapidly identify two species in group G streptococci. The results showed that the number of S. dysgalactiae subsp. equisimilis was twice as many as S. anginosus, and was more invasive than S. anginosus through specimen data analysis. Due to the reports that some specific emm types are responsible for the invasion of GGS, and the recurrent infection of 14 patients, we wanted to connect the relationships between specific emm types and diseases through emm typing and PFGE genotyping. We found out that almost all the S. dysgalactiae subsp. equisimilis were emm typable (with one exception), and about 11.27% were stG6.1 or stG485 respectively, while only 0.85% (n=13) S. anginosus are emm detectable. Meanwhile, recurrent patients were mostly infected by stG485 or stG6.1 S. dysgalactiae subsp. equisimilis. Thus, we could conclude that some specific types are highly responsible for recurrence and invasive disease. In order to understand the drug susceptibility of group G streptococci, we tested with 18 antibiotics and analyzed drug resistance genes. About 20% isolates resistant to macrolide and streptogramin B, and 22% isolates resistant to lincosamide. The low MIC ranges of other antibiotics reveal that clinical drug choice is still working. After double disc testing, 2 out of 50 isolates (4%) were inducible MLSB, 96% displayed M phenotype, and 14 isolates were constitutive MLSB. To further illustrate the macrolide resistance gene, erm(B), erm(TR), mef(A) were tested. Among 64 isolates, 7.8% had erm(B), 4.7% had erm(TR), 75% had mef(A) gene, and 3.1% had both erm(B) and mef(A). At last, because erm(T) gene in S. bovis was flanked by IS1216, we try to find out whether GGS had similar structure. The results of long-term PCR, TA cloning and sequencing shows that there was a tet(S) gene, instead of erm(T), inside the IS1216. This structure was similar to Lactocococcus lactis subsp. lactis pk214. PCR screening reveals 12 isolates displayed IS1216-tet(S) fragment, and southern blotting shows that there was only a single copy of this structure. Also, the PFGE of Apa I could tell us the clonal spreading of two kinds of IS1216-tet(S) models. PFGE of I-Ceu I confirmed the chromosome position of IS1216-tet(S), and the Long-Accurate PCR sequencing shows the upstream and downstream genes of the three different truncated models. The horizontal transfer of the IS1216- tet(S) through conjugation was not detected, however, we assumed that transposon or plasmid with the IS1216-tet(S) prototype was once largely disseminated, and might have different level of deletion during the process of chromosome insertion.