Chester Clinic

 

Microsoft word - iuf.sib.shiga toxin producing e. coli-german outbreak.doc

IUFoST Scientific Information Bulletin (SIB)
June 2011
SHIGA TOXIN PRODUCING Escherichia coli: GERMANY 2011 Escherichia coli
O1O4:H4 OUTBREAK LINKED TO SPROUTED SEEDS

Introduction
The outbreak of Shiga Toxin producing Escherichia coli O104:H4 linked to bean sprouts led to over 3800
confirmed cases of illness that included more than 823 cases of Heamolytic Uremic Syndrome (HUS) and
44 deaths (Frank et al., 2011). Although it appeared that the O104:H4 serotype came from nowhere the
pathogen had actually been isolated in Germany 10 years ago. The outbreak in Germany was more a
consequence of two vulnerabilities (highly virulent pathogen and a high risk food) within the food chain
coming together with disastrous consequences.
The purpose of this Scientific Information Bulletin (SIB) is to provide background into Shiga Toxin
Escherichia coli and how the O104:H4 serotype has changed our understanding of pathogenicity of E.
coli. The future challenges in controlling STEC and research needs will also be discussed.
Escherichia coli
E. coli is encountered in the gastrointestinal tract of all warm blooded animals including humans. The
bacterium has a close relationship with mammals that was established some 140 million years ago when
the first rodents appeared on the Earth. The vast majority of E. coli are non-pathogenic and actually are
essential for health. For example, E. coli produces Vitamin K that we use as part of cell repair (blood
clotting) and are far more effective probiotics than lactic acid bacteria in out-competing would be
pathogens such as Salmonella and Clostridium difficile. E. coli has also found utility as an indicator for
fecal contamination in foods and water. Its presence indicates the potential presence of virulent
pathogens such as Salmonella, E. coli O157 and Shigella, amongst others. Finally, E. coli has been the
workhorse of molecular biology and provided a wealth of information on the function of genes and how
these can be manipulated to produce valuable products such as insulin used to treat diabetes. One of the
main features that makes E. coli easy to manipulate in genetic engineering is the ease with which genetic
material can be introduced and alter the physiological traits of the bacterium. Indeed, it is the acquisition
of genes that gave rise to a sub-set of E. coli termed pathogenic E. coli. The pathogenic E. coli can be
sub-grouped on the mode by which they cause illness and are classified as Enteropathogenic E. coli
(EPEC), Entero-toxigenic E. coli (ETEC), Entero-Diffuse E. coli (EDEC) and Entero Invasive E. coli
(EIEC). The additional two groups of relevance here are Entero Aggregative E. coli (EAggEC) and Shiga
Toxin Producing E. coli.
Shiga Toxin Producing Escherichia coli
Escherichia coli harboring shiga toxin(s) genes collectively fall with the STEC group and encompasses
over 200 different serotypes (Couturier et al.. 2011; Lindqvist et al., 2011; Mathusa et al., 2010). The
different serotypes making up the STEC group were known since the 1980ʼs but apart from O157 were
only considered significant from an academic interest in terms of bacterial evolution given that most
showed little or no virulence (Pennington, 2010). Indeed, over 70% of STEC are considered low risk and
rarely associated with illness. However, a sub-group of non-STEC have increasingly been associated with
serious illness called Heamolytic Uremic Syndrome (kidney failure) that has hitherto being mainly
restricted to the E. coli O157:H7 serotype.
How does STEC cause illness?
The ability of the STEC serotype to cause illness is dependent on the complement of virulence factors.
When E. coli O157 is ingested it can survive the acid of the stomach and eventually reaches the colon
where it binds to a receptor on the epithelial cells of the gastro-intestinal tract. The E. coli O157:H7 then
produces a protein called intimin (encoded by eae gene) and Tir (encoded by tir gene) that facilitates
attachment to the epithelial cell surface. E. coli O157:H7 also forms a tube (Type III secretion system) to
enable proteins to be transferred from the bacteria into the host. Although there are 25 proteins
transferred from E. coli O157:H7 into the epithelial cell the most significant is the Shiga like toxin
(encoded by stx gene). The shiga toxin acts to cut an adenine unit from the ribosomal RNA (rRNA) of the
host. Although this may appear a trivial alteration to RNA structure it is sufficient to stop protein synthesis
thereby resulting in epithelial cell death. E. coli O157:H7 also produces a heamolysin that dissolves blood
cells thereby contributing to the virulence of the pathogen.
When E. coli O157:H7 cells are ingested, the infected person will start developing the common
characteristics of foodborne illness such as fever, nausea, stomach cramps and diarrhea. In healthy
individuals the immune system will neutralize the E. coli O157:H7 thereby removing the infection from the
body (Pennington, 2010). However, in susceptible individuals (young, old and immuno-compromised) the
E. coli O157:H7 can continue to proliferate and the shiga-toxin eventually enters the blood stream and
targets the kidneys causing Hemolytic Uremic Syndrome (HUS). Blood appears in the urine and diarrhea
and eventually leads to kidney failure that can be fatal. Thrombotic thrombocytopenic purpura (TPP) can
also occur whereby patients develop blood clots. The elderly are more prone to TPP and like HUS it can
be potentially fatal (George, 2009; Karpman, 2008; Sanchez et al., 2010; Tarr, 2009).
It has long been considered that the virulence of O157 could be attributed to the presence of the shiga
toxin (stx2), attachment protein (eae) and heamolysin. With non-O157 STEC the virulence factors are
missing or non-functional. In addition the less potent form of the shiga toxin (stx1) is produced.
Consequently, the majority of non-O157 STEC have low pathogenicity towards humans or give mild
symptoms similar to 24 hour flu (viral gastroenteritis).
Emergence of virulent non-O157 STEC
Within the last 5 years there has been an increased incidence of HUS linked to non-O157 STEC although
E. coli O157:H7 still remains the main pathogen of concern. In the US the incidence of O157 STEC is
estimated at 0.9 cases per 100, 000 population that compares with 0.5 for non-O157 STEC (Kasper et al.,
2010). This translates to an estimated 73,000 cases of E. coli O157 infections accounting for 61 deaths.
In comparison, non-O157 STEC infections accounts for 36, 000 cases leading to 30 deaths. The
incidence of HUS from non-O157 STEC infections is <2% that compares to 8% in the case of E. coli
O157.
"Top 6" STEC
To address the increased food safety risk of non-O157 STEC it was proposed by the USDA Food Safety
Inspection Service (FSIS) to classify all STEC as adulterants (i.e., zero tolerance as with E. coli O157). In
effect this meant that meat processors would be required to screen for STEC and if positive issue a
product recall or send the sample for re-processing. However, both regulators and industry acknowledged
that not all STEC serotypes represented the same food safety risk and a large proportion of serotypes
encountered in foods had no or low virulence. In addition, no reliable methods were available for
differentiating the non-pathogenic STEC from those that can cause HUS. Therefore, a compromise was
reached whereby focus would be placed on those STEC that had been previously implicated in cases of
HUS. From the incidence of non-O157 STEC reported to date there are six serotypes that have been
commonly linked to clinical infections. The "Top 6" serotypes (O26, O111, O103, O121, O45 and O145)
accounted for non-O157 STEC cases that resulted in HUS (Kasper et al., 2010). A common feature of the
Top 6 non-O157 STEC is the expression of stx2 and eae thereby fitting the theory that virulent strains
have the full complementary of factors required to cause the fatal HUS condition (Gill and Gill, 2010).
Therefore, diagnostic tests for screening for the presence of the Top 6 STEC are based on presence of
stx2 and eae with absence of either being interpreted as a negative result. The method was AOAC
approved and it is anticipated that the Top 6 will be classed as adulterants by the end of 2011.

Alternative virulence factors present in STEC
The general theory that STEC required stx2 and eae was accepted although failed to explain why certain
serotypes could cause HUS in the absence of eae. For example, in 1993 there was a small cluster of
three HUS cases linked to serotype O113:H21 that was stx2 positive but eae negative (Paton, 1994). A
strain of E. coli O104 (stx2 positive, eae negative) that was related to the 2011 outbreak strain caused a
single case of HUS in Germany in 2001. The ability to cause HUS in the absence of attachment protein
was proposed to be through hyper production of shiga toxin and an alternative virulence factor called
subtilase (a toxin that shuts down the protein factory of the host) (Biscola et al., 2011; Bosilevac et al.,
2011; Lin et al., 2011; Schaffzin et al., 2011). The origins of subtilase remains unclear although it has
genetic sequences related to the virulent pathogens Bacillus anthracis (causes anthrax), Salmonella typhi
(typhoid causing) and Yersinia pestis (plague). The toxin is part of the AB5 toxin family along with shiga
toxin, cholera toxin and ricin (Paton et al., 2004).
Escherichia coli O104:H4
E. coli O104 is a non-O157 STEC that has been rarely implicated in foodborne illness outbreaks. There
was an outbreak of O104:H21 centered in Helena, Montana, USA, linked to fecally contaminated
pasteurized milk (Anon, 1995). The outbreak resulted in 17 cases 67% of which were female with a
median age of 35. Four cases of HUS resulted from the outbreak with no deaths. Serotype O104:H4 was
isolated in Africa in 2001 from patients with diarrhea and again from a woman in Korea who had
contracted HUS. A single case of HUS within Germany in 2001 was attributed to a strain of O104
although it was not recovered again until the 2011 outbreak.
The E. coli O104:H4 implicated in the outbreak centered in Germany was fully genetically characterized
within a week of being identified as the causative pathogen (Anon, 2011; Bielaszewska et al., 2011).
The description of E. coli O104:H21 implicated in the outbreak is as follows:-
Shiga toxin 1 negative
Shiga toxin 2 (stx2a) positive
Intimin (eae gene) negative
Enterohemolysin negative
EAEC (enteroaggregative E. coli) virulence plasmid:
aatA positive (ABC-transporter protein gene)
aggR positive (master regulator gene of Vir-plasmid genes)
aap positive (secreted protein dispersin gene)
agg positive (AAF/I-fimbral subunit-gene)
aggC positive (AAF/I-fimbral operon-gene)
Antimicrobial resistance profile: Resistant to ampicillin, amoxicillin/clavulanic acid,
piperacillin/sulbactam, piperacillin/tazobactam, cefuroxime, cefuroxime-axetil, cefoxitin,
cefotaxime, cetfazidime, cefpodoxime, streptomycin, nalidixic acid, tetracyclin,
trimethoprim/sulfamethoxazol.
From the genes identified within the outbreak strain there was found to be 93% similarity to another type
of pathogenic E. coli referred to as Entero Aggregative E. coli (EAggEC) with the presence of the stx2
linking the serotype to the EHEC group.

Entero Aggrative Escherichia coli
EAggEC is commonly responsible for infant diarrhea primarily in developing countries and also travelers
to developing countries (i.e., Travelers Diarrhea) (Morabito et al., 1998). There have been outbreaks
within industrial countries where EAggEC has been responsible for sporadic diarrhea although is rare
(Huang et al., 2006). The symptoms of EAggEC infection comprises watery diarrhea, occasionally with
blood and mucus lasting 7 – 14 days but the condition is non-lethal provided the patient remains
hydrated.
EAggEC strains exhibit a spectrum of pathogenicity that is dependent on the presence of virulence factors
required to cause illness. E. coli within the EAggEC group harbor a 60MDa plasmid (pAA) that encodes
for surface appendages (fimbria AAF/I, AAF/II and AAF III) and genes encoding for cell-associated
enterotoxins (EAST1 and serine protease) are also found on the pAA plasmid regulated by aggR
(Zamboni et al., 2004).
The fimbria on the surface of EAggEC act to attachment between cells and the host mucosa. The
bacterium forms a "brick-like" structure that essentially forms a biofilm on the lining of the gastro-intestinal
tract thereby providing firm attachment (Elias et al., 1999). The cell associated toxins induce dilation of
crypt openings leading to fluid accumulation in the lumen manifested by the observed perfuse diarrhea.
The enterotoxin encoded by EAST-1 is not restricted to EAggEC but is also present in Enterotoxigenic
and Enteropathogenic E. coli (also associated with Travelers Diarrhea). Some EAggEC strains express
heamolysin, common in E. coli that cause urinary tract infections.
Although EAggEC do not express shiga toxins the genes it has been noted that several strains harbor the
genes or more correctly the lysogenised bacteriophages (Iyoda et al., 2000). E. coli O157:H7 acquired
the shiga toxin phenotype via bacteriophages and it is likely that E. coli O104:H4 also acquired the toxin
via the same mode. The "shiga toxin producing" bacteriophages are lysogenic that integrate stx into the
chromosome of the E. coli host. When the E. coli is subjected to stress (for example, presence of
antibiotics) the SOS response is induced that triggers the replication cycle of the phages, including
expression of shiga toxin that is released when the cell lyses (Turner, 2011). This is the primary reason
why antibiotics are not administered when treating HUS patients. The over use of antibiotics, is leading to
an increase prevalence of stx carrying bacteria is considered a possibility.
Serotype O104:H4 was also found to express ESBL (Extended Spectrum Beta lactamase) which is an
enzyme that degrades antibiotics such as penicillin. ESBL is commonly encountered in urinary tract
infections caused by E. coli.
What was the likely source of E. coli O104:H4 in the 2011 German outbreak?
The outbreak of E. coli O104:H4 was eventually traced to bean spouts produced by an organic farm
outside Hamburg (Frank et al., 2011) Sprouted seeds have been implicated in over 40 outbreaks of
foodborne illness within the last decade (Warriner et al., 2009). The traditional source of contamination
associated with sprouts is the seed although contamination introduced during the sprouting process has
also been implicated in a small number of cases. The seed originally becomes contaminated in the field
and can be exposed to a wide range of hazards derived from manure, contaminated irrigation water, in
addition to wildlife. Pathogens can reside on seeds in a dormant state and rapidly multiply under the warm
(20-25°C) and humid (>90% relatively humidity) conditions of the sprouting room. It has been
demonstrated that pathogens such as E. coli can rapidly multiply from 1 cell to attain 100, 000, 000 cells
per g within 48h of the sprouting period (Warriner et al., 2003). In addition, pathogens can readily become
internalized within sprouting seeds and cannot be removed by washing.
E. coli O157:H7 outbreaks are commonly linked to meat (ground beef) and leafy greens such as spinach.
In contrast, non-O157 STEC have a tendency to be linked to person-to-person contact or food
service/community locations. Furthermore, whilst O157 STEC primarily affects the young and old, non-
O157 STEC appeared to cause fatal HUS over a broader age range especially young adults. In the case
of O104:H4, EAggEC are exclusively linked to human sources given that no animal reservoirs have been
identified to date. If the contamination was introduced onto the seed during cultivation via a sewage spill
or transfer to the sprouting seed bed during sprouting remains to be confirmed.

Conclusions
The E. coli O104:H4 outbreak linked to bean sprouts was one of the largest foodbourne illness outbreaks ever recorded. Over 3000 cases were confirmed with 38 deaths and an unusually high incidence of HUS (30%) that primarily affected young adult females. The outbreak underlines the food safety issues linked to sprouted seeds that have been responsible for over 40 outbreaks over the last decade. In addition, there was the emergence of an apparent new EHEC and EAggEC hybrid in the form of E. coli O104:H4. However, there have been previous cases of HUS linked to serotypes lacking the eae attachment protein although these have been relatively rare. It is likely that in the German outbreak the O104:H4 serotype proliferated to high levels during the sprouting process thereby leading to those affected receiving a high dose of the pathogen. It could be that ordinarily the population is exposed to low doses of the pathogen and hence is rarely encountered. Still there are many features of O104:H4 that need to be known in order to enable control of the serotype. For example, what is the prevalence of the bacterium within the population and why are women more susceptible. Possible theories for the latter includes the administration of birth control pills increasing the receptors required for O104:H4 attachment and/or the involvement of the pathogen in urinary tract infections. Also there is a need to understand how the stx2 gene was transferred to O104:H4 and if there is a risk of the toxin becoming more widely distributed amongst E. coli. Current evidence would suggest that antibiotics may have a direct role in the mobility of the stx gene between strains although that needs to be confirmed. The question of whether more outbreaks of O104:H4 will occur in the future remains open to speculation. The fact that the pathogen is more likely to reside in humans as opposed to animal reservoirs would suggest that the prevalence of O104:H4 would be lower than that of O157 that is frequently encountered in cattle. The infectious dose of the O104:H4 serotype will also need to be determined given that this would contribute to the risk associated with the pathogen. Regardless of this factor the question is posed whether a further subgroup of STEC should be established to account for non-O157 serotypes that belong to the EAggEC and that would be missed in standard assays based on detecting stx2 and eae genes. References
Anon, (2011). Outbreak of life-threatening Haemolytic uremic syndrome (HUS) caused by bacterial infection. Robert Koch Institute http://www.rki.de/EN/Home/homepage node.html? nnn=true Bielaszewska, M., Mellmann, A., Zhang, W., Kock, R., Fruth, A., Bauwens, A., Peters, G. and Karch H (2011). Characterisation of the Escherichia coli strain associated with an outbreak of haemolytic uraemic syndrome in Germany, 2011: a microbiological study. The Lancet Infectious Diseases, doi:10.1016/S1473-3099(11)70165-7 http://www.thelancet.com/journals/laninf/article/PIIS1473-3099(11)70165-7/fulltext?_eventId=login Biscola,F.T., Abe,C.M. and Guth,B.E. (2011).Determination of adhesin gene sequences in, and biofilm formation by, O157 and non-O157 Shiga toxin-producing Escherichia coli strains isolated from different sources.Applied and Environmental Microbiology (77), 2201-2208. Bosilevac,J.M. and Koohmaraie,M. (2011).Prevalence and characterization of non-O157 shiga toxin-producing Escherichia coli isolates from commercial ground beef in the United States.Applied and Environmental Microbiology (77), 2103-2112. http://www.ncbi.nlm.nih.gov/pubmed/21257806 Couturier,M.R., Lee,B., Zelyas,N. and Chui,L. (2011a).Shiga-toxigenic Escherichia coli detection in stool samples screened for viral gastroenteritis in Alberta, Canada.Journal of Clinical Microbiology (49), 574-578. Elias, W. P., Suzart, S., Trabulsi, L. R., Nataro, J. P. and Tania, A (1999). Distribution of aggA and aafA gene sequences among Escherichia coli isolates with genotypic or phenotypic characteristics, or both, of enteroaggregative E. coli. J Med Microbiol 48; 597-599. Frank C., Werber, D., Cramer, J. P., Askar, M., an der Heiden, M., Barnett, H., Fruth, A., Prager, R., Spode, A., Wadj, M., Zoufaly, A., Jordan, S., Stark, K. and Krause, G. (2011). Epidemic Profile of Shiga-Toxin–Producing Escherichia coli O104:H4 Outbreak in Germany — Preliminary Report. New England J Med., 10.1056/NEJMoa1106483 http://www.nejm.org/doi/full/10.1056/NEJMoa1106483?query=featured_home&&&#t=articleDiscussion George, J.N. (2009).The thrombotic thrombocytopenic purpura and hemolytic uremic syndromes: overview of pathogenesis (Experience of The Oklahoma TTP-HUS Registry, 1989-2007).Kidney International. Supplement (112), S8-S10. Gill,A. and Gill,C.O. (2010).Non-O157 verotoxigenic Escherichia coli and beef: a Canadian perspective.Canadian Journal of Veterinary Research = Revue Canadienne De Recherche Veterinaire (74), 161-169. Huang, D. B., Nataro, J. P., DuPont, H. L., Kamat, P. P., Mhatre, A. D., Okhuysen, P. C., Chiang, T. (2006). Enteroaggregative Escherichia coli Is a Cause of Acute Diarrheal Illness: A Meta-Analysis. Clinical Infectious Diseases 43: 556-563 Iyoda S, Tamura K, Itoh K, Izumiya H, Ueno N, Nagata K, Togo M, Terajima J, Watanabe H. (2000). Inducible stx2 phages are lysogenized in the enteroaggregative and other phenotypic Escherichia coli O86:HNM isolated from patients. FEMS Microbiol Lett. 191(1):7-10. Kasper, C, Doyle M. E. and Archer, J (2010). White paper on non-O157:H7 shiga toxin producing E coli from meat and non-meat sources. FRI Food Safety Reviews. http://fri.wisc.edu/docs/pdf/FRI_Brief_NonO157STEC_4_10.pdf Karpman,D. (2008).Hemolytic uremic syndrome and thrombotic thrombocytopenic purpura. Current perspectives on EHEC, complement mutations and ADAMTS13.Lakartidningen (105), 1096-1101. Kumar, M, Hora, R. Kostrzynska, M., Waites W. M. and Warriner K. (2006). Inactivation of Escherichia coli
O157:H7 and Salmonella on Mung Beans, Alfalfa and other Seed Types Destined for Sprout Production
by Using an Oxychloro-Based Sanitizer. J. Food Prot. 69 (7): 1571-1578
Lin,A., Sultan,O., Lau,H.K., Wong,E., Hartman,G. and Lauzon,C.R. (2011).O serogroup specific real time PCR assays for the detection and identification of nine clinically relevant non-O157 STECs.Food Microbiology (28), 478-483. Lindqvist,R. and Lindblad,M. (2011).Time to growth and inactivation of three STEC outbreak strains under conditions relevant for fermented sausages.International Journal of Food Microbiology (145), 49-56. Mathusa,E.C., Chen,Y., Enache,E. and Hontz,L. (2010a).Non-O157 Shiga toxin-producing Escherichia coli in foods.Journal of Food Protection (73), 1721-1736. http://www.ncbi.nlm.nih.gov/pubmed/20828483 Morabito, S., Karach, H., Mariani-Kurkdijan, P., Schmidt, H., Minelli, F., Bingen, E and Caprioli (1998). Enteroaggregative. Shiga toxin producing Escherichia coli O111:H2 associated with an outbreak of hemolytic uremic syndrome. J Clim Microbiol 36; 840-842. http://www.ncbi.nlm.nih.gov/pubmed/9508328 Paton,A.W., Srimanote,P., Talbot,U.M., Wang,H. and Paton,J.C. (2004).A new family of potent AB(5) cytotoxins produced by Shiga toxigenic Escherichia coli.The Journal of Experimental Medicine (200), 35-46. Pennington,H. (2010).Escherichia coli O157.Lancet (376), 1428-1435. Sanchez,S., Martinez,R., Alonso,J.M. and Rey,J. (2010).Clinical and pathogenic aspects of infections due to Escherichia coli O157:H7 and other verocytotoxigenic E. coli.Enfermedades Infecciosas y Microbiologia Clinica (28), 370-374. Schaffzin,J.K., Coronado,F., Dumas,N.B., et al. (2011).Public health approach to detection of non-O157 Shiga toxin-producing Escherichia coli: summary of two outbreaks and laboratory procedures.Epidemiology and Infection 1-7. Tarr,P.I. (2009).Shiga toxin-associated hemolytic uremic syndrome and thrombotic thrombocytopenic purpura: distinct mechanisms of pathogenesis.Kidney International.Supplement ((112)), S29-32. Turner, M (2011). Phage on the rampage; Antibiotic use may have driven the development of Europes deadly E. coli. Nature http://www.nature.com/news/2011/110609/full/news.2011.360.html?s=news_rss Warriner, K, Dunfield K, Namvar A and Hubard, A (2009). Sources and Interaction of Human Pathogens with Fresh Produce. Advances in Food and Nutrition Research 57; 275-301. Warriner, K., Spanioias, S., Dickinson, M., Wright, C. and Waites, W.M. (2003). Internalization of bioluminescent Eschericha coli and Salmonella Montevideo in growing mung bean sprounts. J Appl Microbiol 95; 719-727. Zamboni, A., Fabbricotti. S. H., Fagundes-Neto, U. and Scaletsky, I. A. C. (2004). Enteroaggregative Escherichia coli virulence factors are found to be associated with infantile diarrhea in Brazil. J Clin Microbiol 42; 1058-1063. Prepared by Keith Warriner on behalf of, and and approved by, the IUFoST Scientific Council. Dr Warriner ([email protected]) is a member of the faculty in the Department of Food Science and the current director of the Food Safety and Quality (FSQA) MSc program within the University of Guelph (http://www.uoguelph.ca) The International Union of Food Science and Technology (IUFoST) is the global scientific organisation representingover 200,000 food scientists and technologists from more than 65 countries. It is a federation of national food science organisations linking the world's food scientists and technologists. IUFoST has four regional groupings: ALACCTA representing Central and South America, EFFoST representing Europe, WAAFoST representing Western Africa and FIFSTA representing the countries in the ASEAN region.IUFoST Contact: J. Meech, Secretary-General, IUFoST, P O Box 61021, No. 19, 511 Maple Grove Drive, Oakville,Ontario, Canada, L6J 6X0, Telephone: + 1 905 815 1926, Fax: + 1 905 815 1574, e-mail: [email protected]

Source: http://www.foodsecurity.or.kr/bbs/download.php?&bbs_id=qnaa05&page=&type=1&doc_num=24