Introduction: Qualified Presumption of Safety QPS

Read Time: 40 minutes | My Lab Experts

SUMMARY 

A wide variety of microbial species are used in food and feed production. Some have a long history of apparent safe use. The FDA have a classification for products that are generally consider safe,(GCS) and “Qualified Presumption of Safety (QPS)”. This product meets the criteria for both. The probiotic cells in this formula are from the Lactobaccilus plantarum family.

The EFSA Journal (2007) 587, 1-16  - Introduction of a Qualified Presumption of Safety (QPS) approach for assessment of selected microorganisms referred to EFSA1 - Opinion of the Scientific Committee  -(Question No EFSA-Q-2005-293) 

Adopted on 19 November 2007 

SCIENTIFIC COMMITTEE MEMBERS Sue Barlow, Andrew Chesson, John D. Collins, Erik Dybing, Albert Flynn, Claudia Fruijtier- Pölloth, Anthony Hardy, Ada Knaap, Harry Kuiper, Pierre Le Neindre, Jan Schans, Josef Schlatter, Vittorio Silano, Staffan Skerfving, Philippe Vannier. 

 

To meet this need a system was proposed for a pre-market safety assessment of selected groups of microorganisms leading to a “Qualified Presumption of Safety (QPS)”. In essence this proposed that a safety assessment of a defined taxonomic group (e.g. genus or group of related species) could be made based on four pillars (establishing identity, body of knowledge, possible pathogenicity and end use). If the taxonomic group did not raise safety concerns or, if safety concerns existed, but could be defined and excluded (the qualification) the grouping could be granted QPS status. Thereafter, any strain of microorganism the identity of which could be unambiguously established and assigned to a QPS group would be freed from the need for further safety assessment other than satisfying any qualifications specified. Microorganisms not considered suitable for QPS would remain subject to a full safety assessment. 

EFSA asked its Scientific Committee to consider whether this system could be used to harmonise approaches to the safety assessment of microorganisms across the various EFSA scientific panels. If so, the Committee was requested to develop a strategy for the introduction of an assessment system based on the QPS concept. 

For citation purposes: Opinion of the Scientific Committee on a request from EFSA on the introduction of a Qualified Presumption of Safety (QPS) approach for assessment of selected microorganisms referred to EFSA. The EFSA Journal (2007) 587, 1-16 

© European Food Safety Authority, 2007 

The Scientific Committee reviewed the range and numbers of microorganisms likely to be the subject of an EFSA opinion. They found that approximately 100 species of microorganisms have been or are expected to be referred to EFSA for a safety assessment; the majority being the result of notifications for market authorization as sources of food and feed additives, food enzymes and plant protection products. A large majority of these species were found to fall within four broad groupings: i) Gram-positive non-sporulating bacteria; ii) Bacillus species, iii) yeasts and iv) filamentous fungi. Accordingly, bacteria, yeasts and fungi falling within these four groups were selected for an initial assessment of their suitability for QPS status. The Scientific Committee concluded that the weight of evidence available for many species falling within the first three of the four groups was sufficient to ensure that QPS status provided at least the same degree of confidence as a case-by-case safety assessment. However, the Committee found that, in the case of the filamentous fungi, the body of knowledge, particularly that relating to a history of use, was for a specific purpose and did not allow extrapolation to other uses to be made with confidence and so could not recommend QPS status for such fungi. 

As the number of organisms considered suitable for QPS status is sufficiently extensive to cover a majority of the safety assessments involving microorganisms required of EFSA, the Scientific Committee concluded that the introduction of a QPS system for microorganisms would meet the objectives of providing a practical tool for setting priorities and avoiding the extensive investigations of organisms known not to cause concern. Although QPS status of metabolic products of microorganisms cannot be inferred from the QPS status of the production strain, the Committee considered that the system still had value for the assessment of strains used in the production of such products. Further work, however, would be required to extend the system to encompass those microorganisms used for biological control purposes. 

Finally, in reaching its conclusion on the value of QPS as an assessment tool, the Scientific Committee recognized that there would have to be continuing provision for reviewing and modifying the list of organisms given QPS status. They recommended that the EFSA via its Science Directorate should take prime responsibility for this and should review the suitability for QPS status of the existing list and any additions at least annually. Reviews may occur more frequently as necessary but there should be a formal requirement that even when no changes are proposed, a statement should be made annually that QPS status is being maintained for the published list. 

Key words: 

Safety assessment, microorganisms, qualified presumption of safety, QPS, Bacillus, yeast, filamentous fungi, Gram-positive non-sporulating bacteria, lactic acid bacteria 

Qualified Presumption of Safety (QPS) 

The EFSA Journal (2007) 587, 2-16 

Qualified Presumption of Safety (QPS) 

The EFSA Journal (2007) 587, 3-16 

BACKGROUND 

A wide variety of bacterial and fungal species are used in food and feed production, either directly or as a source of additives or food enzymes. Some of these have a long history of apparent safe use, while others are less well understood and may represent a risk for consumers. Experience has shown that there is a need for a tool for setting priorities within the risk assessment of those microorganisms used in the production of food/feed which are captured by present legislation and consequently the subject of a formal assessment of safety. Ideally such an assessment tool would allow the identification of risk without committing resources to extensive investigations of organisms known to be safe. 

In 2002/3 a working group consisting of members of the former Scientific Committees on Animal Nutrition, Food and Plants of the European Commission proposed the introduction for selected microorganisms of a Qualified Presumption of Safety (QPS)2. 

In essence this proposed that a safety assessment of a defined taxonomic group (e.g. genus or group of related species) could be made independently of any particular pre-market authorization process. If the taxonomic group did not raise safety concerns or, if safety concerns existed but could be defined and excluded (the qualification) the grouping could be granted QPS status. Thereafter any strain of microorganism the identity of which could be unambiguously established and assigned to a QPS group would be freed from the need for further safety assessment other than satisfying any qualifications specified. Those strains failing to satisfy a qualification would be considered hazardous and, in the absence of mitigating circumstances, unfit for purpose. Microorganisms not considered suitable for QPS would remain subject to a full safety assessment. 

In April 2003, responsibility for the safety assessments of food/feed undertaken by the Scientific Committees of the Commission formally passed to the European Food Safety Authority (EFSA). Shortly after EFSA asked its own Scientific Committee to consider whether the approach to safety assessment of microorganisms proposed in the QPS document could be used to harmonise approaches to the safety assessment of microorganisms across the various EFSA scientific panels. In doing so, the Committee was requested to take into account the response of the stakeholders to the QPS approach. Their views had been sought by the three Commission Scientific Committees in 2002/3 and, subsequently, by EFSA at a Scientific Colloquium organised at the end of 2004 (EFSA 2005b). 

The Scientific Committee concluded that QPS as a concept could provide a generic assessment system for use within EFSA that could be applied to all requests received for the safety assessments of microorganisms deliberately introduced into the food chain (EFSA 2005a). The benefits of the introduction of QPS would be a more transparent and consistent approach across the EFSA panels and the potential to make better use of resources by focussing on those organisms which presented the greatest risks or uncertainties. 

Qualified Presumption of Safety (QPS) 

The EFSA Journal (2007) 587, 4-16 

However, the Committee stressed that the body of knowledge about the organisms for which QPS is sought must be sufficient to provide adequate assurance that any potential to produce adverse effects in humans, livestock or the wider environment is understood and predictable. Judgement as to whether the existing data are sufficient needed, in the view of the Committee, to be determined by an expert group established for this purpose and should be based on a weight-of-evidence approach. 

On the basis of these conclusions the Scientific Committee recommended that EFSA should develop a strategy for the introduction of an assessment system based on the QPS concept. This should be limited to microorganisms introduced into the food chain or used as producer strains for food/feed additives until the robustness and value of such a system could be tested in practice. 

EFSA accepted the recommendation of its Scientific Committee and proposed that the Committee should continue its assessment of the QPS system with a view to implementation3. Specifically, the Scientific Committee was asked first to establish which were the microorganisms most commonly referred to EFSA, including those used as a source of microbial products. Then, on the basis of this survey, to select relevant groups of microorganisms, examine the available data on safety and propose whether QPS status would be appropriate. If this proved possible in a significant number of cases then the Scientific Committee should consider how implementation of QPS across the various panels could be achieved. 

TERMS OF REFERENCE AS PROVIDED BY EFSA 

In response to its Opinion on the potential value of the QPS approach, the Scientific Committee is now requested by the European Food Safety Authority: 

1. To establish which are the microorganisms most commonly referred to EFSA. This to include both organisms deliberately introduced into the food chain and those used as a source of microbial products entering the chain. 
2. To select appropriate and relevant groups of microorganisms and, with the help of additional experts as necessary, to determine whether QPS status should be given. 
3. Thereafter, to advise whether QPS represents a practical and robust method of safety assessment for microorganisms and, if so, to consider how the QPS could be applied across EFSA within the framework of the current and proposed legislation. 

See http://www.efsa.europa.eu/en/science/sc_commitee/sc_documents/1368.html 

Qualified Presumption of Safety (QPS) 

The EFSA Journal (2007) 587, 5-16 

ACKNOWLEDGEMENTS 

The European Food Safety Authority wishes to thank the members of the Working Group for the preparation of this opinion: Javier Cabanes, Andrew Chesson (Chairman), Pier Sandro Cocconcelli, Manfred Gareis, Per Einer Granum, John Heritage, Günter Klein, Naresh Magan, Bevan Moseley, Christophe Nguyen-The, Fergus Priest, Amparo Querol, Tine Rask Licht, Malcolm Richardson, Robert Samson, Klaus Peter Schaal, Ulf Thrane, Atte von Wright. 

Qualified Presumption of Safety (QPS) 

The EFSA Journal (2007) 587, 6-16 

ASSESSMENT 

1. Questions to EFSA involving microorganisms 

Approximately 100 species of microorganisms have been or are expected to be referred to EFSA for a safety assessment. The majority are the result of notifications for market authorization as sources of food and feed additives, food enzymes and plant protection products. Others are the subject of the GMO and novel food/feed legislation. A few microbial species are also the subject of requests for opinions relating to consumer or animal safety not directly linked to product authorization or to legislative requirements. Generally such requests relate to human enteropathogens or veterinary pathogens and so are beyond the scope of any consideration for QPS. Microorganisms referred to EFSA include both live organisms deliberately introduced into the food chain and those used as a source of food/feed additives and food enzymes. Individual species may be the subject of a single notification but more usually are found in several notifications. A large majority of these approximately 100 species were found to fall within four broad groupings: 

1. Gram-positive non-sporulating bacteria (GPNS) 2. Bacillus species 3. Yeasts 4. Filamentous fungi 

Accordingly, bacteria and fungi falling within these four groups were selected for an initial assessment of their suitability for QPS status. Organisms falling outside the four broad groups are infrequently notified. The Scientific Committee considers that such organisms could be considered for QPS at a later date but, in the interim, should continue to be assessed on a case- by-case basis. This would include viruses which are the occasional subject of notifications. 

It should be noted that QPS status is taken to apply strictly to the microorganism and not to any traded product containing the organism or to a product of the microorganism. The Scientific Committee recognises that the final formulation may, on rare occasions, introduce additional hazards needing assessment. In addition, QPS status informs only on the safety of a microorganism and should be used without prejudice to any other requirements of legislation. 

2. Consideration for QPS status 

The suitability of various taxonomic groups falling under the four broad headings for QPS status was examined by working groups of the Scientific Committee. Their preliminary proposals for suitable candidates for QPS status and the documentation supporting these conclusions were made available for public consultation. Interested parties were invited in particular to comment on whether the weight of evidence presented was sufficient to ensure that QPS status provides at least the same degree of confidence as a case-by-case safety assessment, whether this was adequately documented and whether there were issues that have not been sufficiently considered. 

Qualified Presumption of Safety (QPS) 

The EFSA Journal (2007) 587, 7-16 

Responses from some thirty individuals and organizations were received, principally from trade organizations and companies, but including scientists from a number of European academic institutions, trade associations and national food safety authorities. Relatively few comments were directed to the specifics of the analyses of suitability for QPS status. Most were concerned with more general issues, some positive and others raising matters which were considered insufficiently developed. Concerns remain about the status of QPS in relation to existing and future legislation recognizing that the application of QPS by EFSA could have implications for risk managers. The issue of how a QPS list would be maintained was also raised, since scientific developments might require a species to be withdrawn or allow a species to be added. There were also concerns about the continuing emphasis on the absence of acquired antibiotic resistance determinants as a qualification in bacteria and that a restriction on end-use was not more generally applied to allow additional organisms to be considered suitable for QPS, particularly amongst the filamentous fungi. 

All of the comments received were taken into consideration when reviewing and revising the conclusions on suitability for QPS status. The organisms which the Scientific Committee considers suitable for QPS status is given in Table 1. For convenience this is given as a list of presently-recognized species. Where QPS status is proposed, the Scientific Committee is satisfied that the body of knowledge available is sufficient to provide adequate assurance that any potential to produce adverse effects in humans, livestock or the wider environment is understood and capable of exclusion. 

A summary of some of the specific issues arising within each of the four broad groupings is highlighted below. Otherwise the scientific justifications for inclusion in this list are given in the individual reports on the four groupings (Appendices A-D). These reports take a common structure based around the four pillars of the assessment for suitability for QPS status (establishing identity, body of knowledge, possible pathogenicity and end use) following the general scheme previously published4. The current state of knowledge and the very different nature of the organisms involved, however, have meant that the emphasis and content within the four reports inevitably differs. Each individual report is a summary of extended considerations based on a thorough review of the available scientific literature and the knowledge and experience of the scientists involved. Where literature is cited, this is to support key conclusions or more generally to illustrate an issue. 

See http://www.efsa.europa.eu/etc/medialib/efsa/science/colloquium_series/no2_qps/948.Par.0015.File.dat/summary_report1.pdf , page 16 

Qualified Presumption of Safety (QPS) 

The EFSA Journal (2007) 587, 8-16 

Qualified Presumption of Safety (QPS) 

Table 1. List of taxonomic units proposed for QPS status 

Gram-Positive Non-Sporulating Bacteria5 Species Qualifications Bifidobacterium adolescentis Bifidobacterium animalis 

Bifidobacterium bifidum Bifidobacterium breve 

Bifidobacterium longum 

Corynebacterium glutamicum QPS status applies only when the species is used 

for production purposes. Lactobacillus acidophilus Lactobacillus amylolyticus Lactobacillus amylovorus Lactobacillus alimentarius Lactobacillus aviaries Lactobacillus brevis Lactobacillus buchneri Lactobacillus casei Lactobacillus crispatus Lactobacillus curvatus Lactobacillus delbrueckii 

Lactobacillus farciminis Lactobacillus fermentum Lactobacillus gallinarum Lactobacillus gasseri Lactobacillus helveticus Lactobacillus hilgardii Lactobacillus johnsonii Lactobacillus kefiranofaciens Lactobacillus kefiri Lactobacillus mucosae Lactobacillus panis 

Lactobacillus paracasei Lactobacillus paraplantarum Lactobacillus pentosus Lactobacillus plantarum Lactobacillus pontis Lactobacillus reuteri Lactobacillus rhamnosus Lactobacillus sakei Lactobacillus salivarius Lactobacillus sanfranciscensis Lactobacillus zeae Lactococcus lactis 

Leuconostoc citreum Leuconostoc lactis Leuconostoc mesenteroides Pediococcus acidilactici Pediococcus dextrinicus Pediococcus pentosaceus 

Propionibacterium. freudenreichii Streptococcus thermophilus 

Absence of acquired antibiotic resistance should be systematically demonstrated unless cells are not present in the final product (EFSA, 2005c). 

The EFSA Journal (2007) 587, 9-16 

Table 1 (cont’d). List of taxonomic units proposed for QPS status 

Bacillus6 Species Qualifications Bacillus amyloliquefaciens Bacillus atrophaeus Bacillus clausii Bacillus coagulans Bacillus fusiformis 

Bacillus lentus Bacillus licheniformis Bacillus megaterium Bacillus mojavensis 

Bacillus pumilus Bacillus subtilis Bacillus vallismortis Geobacillus stearothermophillus 

Absence of emetic food poisoning toxins with surfactant activity.* Absence of enterotoxic activity.* 

* When strains of these QPS units are to be used as seed coating agents, testing for toxic activity is not necessary, provided that the risk of transfer to the edible part of the crop at harvest is very low (section 4.3 of Appendix B). 

Yeasts Species Qualifications Debaryomyces hansenii Hanseniaspora uvarum Kluyveromyces lactis Kluyveromyces marxianus 

Pichia angusta Pichia anomala 

Saccharomyces bayanus Saccharomyces cerevisiae Saccharomyces pastorianus (synonym 

of Saccharomyces carlsbergensis) 

1. cerevisiae, subtype S. boulardii is contraindicated for patients of fragile health, as well as for patients with a central venous catheter in place. A specific protocol concerning the use of probiotics should be formulated Schizosaccharomyces pombe Xanthophyllomyces dendrorhous 

Qualified Presumption of Safety (QPS) 

The EFSA Journal (2007) 587, 10-16 

Qualified Presumption of Safety (QPS) 

2.1. Gram-positive non-sporulating bacteria 

Many of the referred microorganisms falling within this grouping are normal inhabitants of the digestive tract of humans and livestock or are commonly used in the preparation of foods and feed. Consequently, there has been a long history of human exposure with only very occasional reports of adverse effects and then only amongst compromised individuals. However, amongst the microorganisms referred to EFSA, two particular groups of microorganisms raised issues requiring particular attention. The most important was the consideration given to the enteroccoci. Bacteria in the genus Enterococcus are amongst the leading causes of community- and hospital-acquired (nosocomial) infections. Infections often result from Enterococcus faecalis, but there are also virulent strains found within E. faecium, the species of Enterococcus most commonly deliberately introduced into the food chain. Although a considerable amount is known about the virulence determinants in enteroccoci, given the prevalence of enteroccocal infections, the Scientific Committee verged on the side of caution and did not propose QPS status for Enterococcus species, as at present it is not possible readily to distinguish between virulent and non-virulent strains without resorting to the level of investigation used in a case- by-case assessment. This position could be reviewed as it becomes clearer which are the key determinants of virulence and as suitable molecular probes for such determinates are developed. 

The second issue highlighted the debate about the distinction between opportunistic infections, of which almost all microorganisms that humans commonly encounter are capable, and pathogenicity. Many Lactobacillus species have been occasionally encountered in clinical specimens, the clinical significance of which is not always clear. Such occurrences have almost invariably been associated with immunocompromised patients, those who had suffered surgical or accidental insult or who had a serious underlying illness, and remain rare. As such, these infections can be considered opportunistic and beyond the capacity of any safety assessment to exclude. Although a number of Lactobacillus spp. have been reported to infect otherwise healthy individuals with a history of rheumatic endocarditis or following heart valve replacement, one species, L. rhamnosus, appears to predominate. This organism could be considered on the edge of being defined as pathogenic. The Scientific Committee took the view that the at-risk population is not placed at added risk by the use of L. rhamnosus in food/feed and so confirmed the proposed QPS status of this bacterium. The Committee considers that this is a decision which should be reviewed at regular intervals. 

2.2. Bacillus species 

The Scientific Committee is of the opinion that the use of strains from the B. cereus group should be avoided whenever there is a possibility of human exposure whether intended or incidental. The B. cereus group is therefore excluded from consideration for QPS status. 

There is an artificial distinction held between B. cereus and B. thuringiensis (used for plant protection) which has little scientific basis. The plasmid encoding the insecticidal enterotoxin, which provides the phenotypic distinction for B. thuringiensis, is readily lost, particularly when grown at 37oC, leaving an organism indistinguishable from B. cereus. Consequently it is likely 

The EFSA Journal (2007) 587, 11-16 

Qualified Presumption of Safety (QPS) 

that B. thuringiensis has been the causative organism of some instances of food poisoning but identified as B. cereus because clinical investigations would have failed to recognise the distinguishing features characteristic of B. thuringiensis. 

However, the Scientific Committee recognises that B. thuringiensis has value to the industry as a means of biological pest control and that its widespread use for this purpose may not lead to significant human exposure. 

Although occasional strains of Bacillus not falling within the B. cereus group also produce human enterotoxins, experience gained with B. cereus has provided the tools for their exclusion. This is recognised as a qualification in recommending QPS status for other Bacillus species. 

2.3. Yeasts 

Yeasts used in food production, particularly brewers/bakers yeast, are considered amongst the safest of microorganisms. However, even amongst this group there are reports of very occasional invasive infections. A sub-type of Saccharomyces cerevisiae, commonly referred to as Saccharomyces boulardii has been used as an adjunct to the antibiotic treatment of persistent diarrhoea often arising from Clostridium difficile infections, to reduce the likelihood of reoccurrence. This subtype has been isolated from the blood in approximately half of the reported invasive infections involving Saccharomyces. The majority of these cases occurred in compromised individuals. However this has to be placed in context. Despite the continuous and universal exposure to this yeast there have been less than a hundred documented cases of invasive infection by Saccharomyces spp., half of which occurred amongst those undergoing aggressive antibiotic treatment. As the at-risk group results from a strictly medical application without implication for the healthy population, the Scientific Committee did not see a reason to exclude Saccharomyces and so confirmed the proposed QPS status of this genus. 

2.4. Filamentous fungi 

The filamentous fungi could not be included within the QPS system. Although in many cases there has been a history of use, this has been for specific purposes such as the production of citric acid or processing enzymes. The body of knowledge that has developed has, in consequence, centred on these uses. However, many of the filamentous fungi used for production purposes are known to produce substances of potential concern (mycotoxins, etc). 

The strength of the QPS lies in the ability to provide a generic system of safety assessment. This can be extended in scope by introducing a limited number of qualifications, allowing the majority of a taxonomic group to be assumed safe while excluding a minority of problematic strains. On examination, while it was possible to identify specific metabolites of filamentous fungi which should be excluded, it was not possible to be sure that these represented the totality of substances of concern capable of being produced by the taxonomic unit. Introducing restricted use as a qualification did not offer a solution since purpose does not offer any reassurance on overall metabolic capacity. The absence of undesirable compounds in one or 

The EFSA Journal (2007) 587, 12-16 

Qualified Presumption of Safety (QPS) 

more selected production strains does not allow extrapolation to all strains within the selected taxonomic unit. 

3. QPS as a tool for the assessment of the safety of microorganisms 

3.1. Value to EFSA of QPS as an assessment tool 

The list of organisms in Table 1 is sufficiently extensive to cover a large majority of the safety assessments involving microorganisms required of EFSA. Consequently, the Scientific Committee considers the introduction of a QPS system for microorganisms would meet the original objectives of providing a practical tool for setting priorities and avoiding the extensive investigations of organisms already known to be safe. 

Although QPS status of metabolic products of microorganisms cannot be inferred from the QPS status of the production strain, the system still has considerable value for the assessment. Microbial products (e.g. enzymes, organic acids, amino acids) used in food/feed are rarely the primary source of possible concerns. Most case-by-case assessments focus on the presence of other metabolites which might be carried through to the final product. The QPS status of the production strain would provide the assurance that any metabolites other than that intended which are found in the final product would not be hazardous. This would simplify and greatly assist the assessment process. 

Genetically modified microorganisms (GMMs) are the subject of specific legislation, whether used directly and released into the environment (Regulation 1829/20036, Directive 2001/18/EC7) or used under containment as a source of specific products (Directive 98/81/EC8). In the simplest case, that of self-cloning, the QPS status of the parent strain should be accepted and extrapolated to the modified strain without further need for assessment (EFSA 2006). Whenever foreign genetic material is introduced in a GMM, QPS is most likely to be of relevance to the recipient strain and only rarely to the source of the introduced trait. When the recipient strain has QPS status, then the assessment is free to focus on the introduced trait(s), relying on the reviewed body of knowledge to exclude potential hazards arising from the recipient strain. 

QPS in its present form does not offer a generic approach to the safety assessment of microorganisms used as biological control agents. Most are based on filamentous fungi or bacteria hazardous to humans when they are directly exposed. Indeed the protection offered to plants by such organisms may depend, in part at least, on these toxic principles. However, the Scientific Committee considers that it may be possible to devise robust use qualifications which would allow a QPS approach in the future. Such qualifications would have to include a consideration of effects on non-target species. 

6 See http://eur-lex.europa.eu/LexUriServ/site/en/oj/2003/l_268/l_26820031018en00010023.pdf 7 See http://eur-lex.europa.eu/LexUriServ/site/en/oj/2001/l_106/l_10620010417en00010038.pdf 8 See http://eur-lex.europa.eu/LexUriServ/site/en/oj/1998/l_330/l_33019981205en00130031.pdf 

The EFSA Journal (2007) 587, 13-16 

Qualified Presumption of Safety (QPS) 

3.2. Exclusion from the QPS list 

Only a positive list of microorganisms judged suitable for QPS status is given in Table 1. Exclusion from this list was for a variety of reasons. Many microorganisms commonly encountered in food production were not considered because they are not presently the subject of pre-market authorisations and so would not be notified to EFSA. Other microorganisms were considered (e.g. the enterococci), but the potential risks associated with their use could not be fully defined or, where recognised, the tools necessary for the exclusion of hazardous strains were considered insufficient. Often the body of knowledge, particularly that relating to a history of use, was for a specific purpose and did not allow extrapolation to other uses to be made with confidence. 

It should be stressed that the absence of a particular organism from the list of microorganisms judged suitable for QPS does not necessarily imply any risk associated with its use. Individual strains may be safe but this cannot be judged from the existing knowledge of the taxonomic unit to which it belongs. Consequently, all microorganisms not considered for QPS or not considered as suitable would remain subject to a full safety assessment. 

3.3. Maintenance of QPS list 

In reaching its conclusion on the value of QPS as an assessment tool, the Scientific Committee recognises that there would have to be continuing provision for reviewing and modifying the list of organism given QPS status. EFSA must be able to respond to any new information such as epidemiological data which might suggest that an inclusion on the list should be reconsidered. Similarly, there must be provision for additions. It is likely that the rapid developments in microbial genomics, the full annotation of genomes and the ability to predict metabolic pathways accurately from such annotation will allow inclusion of organisms not presently listed. Proposals for additions could arise from a variety of source including actual notifications to EFSA or at the request of interested parties. However, whatever the route and source of information on which on a judgement of suitability for QPS status is made, the assessment itself and the final judgement on whether to include or exclude must remain within EFSA’s purlieu. 

The Scientific Committee is of the opinion that responsibility for the maintenance and development of the QPS system should be the responsibility of EFSA via its Science Directorate. This would ensure that the system has the continuing necessary support and that there would be a continuity of approach. The Scientific Committee suggests that the EFSA should review at least annually the received notifications involving microorganisms and the suitability for QPS status of any new additions. The review should also consider suggestions arising from elsewhere (e.g. industry, Member States) for possible additions to or deletions from the QPS list. Reviews may occur more frequently but there should be a formal requirement that even where no changes are proposed a statement should be made annually that QPS status is being maintained for the published list. 

The EFSA Journal (2007) 587, 14-16 

Qualified Presumption of Safety (QPS) 

3.4. Implementation of QPS within EFSA 

For QPS to be effective it must be implemented across EFSA for all safety considerations of microorganisms intentionally added to the food chain, regardless of purpose. There should be full harmonisation and implementation of QPS in all EFSA Panels wherever applicable. 

For those panels concerned with risk assessments leading to market authorisations there will be an interim period after the introduction of the QPS system during which Dossiers will reflect previous requirements and contain data made unnecessary for those organisms listed in Table 1. The Scientific Committee suggests that this should be managed and a consistent response adopted in which QPS status is given priority but accompanied by an acknowledgement that data has been provided which is consistent with the QPS status of the organism(s). During this initial period it is suggested that the Scientific Committee should monitor the introduction of QPS and act as a forum in which any difficulties could be resolved. 

CONCLUSIONS AND RECOMMENDATIONS 

CONCLUSIONS 

The Scientific Committee is of the view that the weight of evidence available for the bacterial and fungal species listed in Table 1 is sufficient to ensure that QPS status provides at least the same degree of confidence as a case-by-case safety assessment, 

As the number of organisms considered suitable for QPS status is sufficiently extensive to cover a majority of the safety assessments involving microorganisms required of EFSA, the Scientific Committee considers the introduction of a QPS system for microorganisms would meet the objectives of providing a practical tool for setting priorities and avoiding the commitment of resources to extensive investigations of organisms known not to cause concern. 

In reaching its conclusion on the value of QPS as an assessment tool, the Scientific Committee recognises that there would have to be continuing provision for reviewing and modifying the list of organism given QPS status. 

RECOMMENDATIONS 

The Scientific Committee recommends that a QPS system for microorganisms should be introduced initially covering the organisms listed in Table 1 and that should be implemented across EFSA for all safety considerations of microorganisms intentionally added to the food chain, regardless of purpose. Thereafter, and based on the experience gained from its use in practice, the extension of the QPS system to microbial products could be explored. 

EFSA should take prime responsibility for the maintenance and development of QPS and should review at least annually the suitability for QPS status of the existing list and/or of any additions. Reviews may occur more frequently as necessary but there should be a formal requirement that even where no changes are proposed a statement should be made annually that QPS status is being maintained for the published list. 

The EFSA Journal (2007) 587, 15-16 

Qualified Presumption of Safety (QPS) 

REFERENCES 

EFSA, 2006. Guidance document of the Scientific Panel on Genetically Modified Organisms for the risk assessment of genetically modified microorganisms and their derived products intended for food and feed use. The EFSA Journal 374: 1-115. 

EFSA, 2005a. Opinion of the Scientific Committee on a request from EFSA related to a generic approach to the safety assessment by EFSA of microorganisms used in food/feed and the production of food/feed additives. The EFSA Journal 226: 1-12. http://www.efsa.europa.eu/en/science/sc_commitee/sc_opinions/972.html 

EFSA, 2005b. Summary report of the EFSA Scientific Colloquium on Qualified Presumption of Safety of micro-organisms in food and feed. Held on 13-14 December 2004 in Brussels, Belgium. ISBN 92-9199-012-4. http://www.efsa.europa.eu/en/science/colloquium_series/no2_qps.html 

EFSA, 2005c. Opinion of the Scientific Panel on Additives and Products or Substances used in Animal Feed on the updating of the criteria used in the assessment of bacteria for resistance to antibiotics of human or veterinary importance. The EFSA Journal 223: 1-12. http://www.efsa.europa.eu/EFSA/Scientific_Opinion/feedap_op_ej223_antibiotics_en1,0.pdf 

The EFSA Journal (2007) 587, 16-16 

APPENDIX A. Scientific report on the Assessment of Gram-Positive Non- 

Sporulating Bacteria 

Many of the species included in this broad grouping are common constituents of the normal gut flora of humans and livestock although their occurance and numbers are host dependent. Additionally, species of Gram-positive bacteria constitute common components of the microbial community of food and, for their relevant role in food fermentation; these microorganisms have been deliberately introduced into food as starter cultures. In addition, several bacterial strains belonging to this group have a long history of apparent safe use as food starter cultures, feed additives (e.g. animal probiotics and silage inoculants) and source of additives (e.g. enzymes and amino acids). Based on their habitat and their extensive application in the food and feed sector, many species were judged potentially suitable for their safety to be assessed by the “Qualified Presumption of Safety” (QPS) methods, according to (EFSA 2005). The following genera, all belonging to the phylum Firmicutes, have been considered: Bifidobacterium, Corynebacterium, Enterococcus, Lactobacillus, Lactococcus, Leuconostoc, Pediococcus, Propionibacterium and Streptococcus. 

1 Bifidobacterium 

Bifidobacteria are part of the normal gut microbiota of adults and are also one of the first genera to colonise the gut of infants. In addition, they are normal inhabitants of the gut of animals. A limited number of Bifidobacterium species have a history of use in dairy products, especially sour milk products like yoghurts and more recently yoghurt and fermented milk drinks. 

1.1 Taxonomic unit defined 

Bifidobacteria belong to the Actinomycetes branch of phylum Firmicutes. They are non-motile, non-sporeforming rods of variable appearance, usually curved and clubbed, and are often branched including Y and V forms. They are normally strictly anaerobic, although some species and strains tolerate oxygen. The type species is Bifidobacterium bifidum. Bifidobacteria are saccharolytic organisms and they have the ability to ferment glucose, galactose and fructose. Glucose is fermented via the fructose-6-phosphate shunt to acetic and lactic acid. Differences occur between species in their ability to ferment other carbohydrates and alcohols. 

The genus consists currently of following species: Bifidobacterium adolescentis, B. angulatum, B. animalis subsp. Animalis, B. animalis subsp. lactis, B. asteroides, B. bifidum, B. boum, B. breve, B. catenulatum, B. choerinum, B. coryneforme, B. cuniculi, B. dentium, B. gallicum, B. gallinarum, B. indicum, B. longum, B. magnum, B. merycicum, B. minimum, B. pseudocatenulatum, B. pseudolongum subsp. globosum, B. pseudolongum subsp. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 1 

pseudolongum, B. psychraerophilum, B. pullorum, B. ruminantium, B. saeculare, B. scardovii, B. subtile, B. thermacidophilum subsp. porcinum, B. thermacidophilum subsp. thermacidophilum, B. thermophilum . 

1.2 Is the body of knowledge sufficient? 

The characteristics and habitat of the species of the genus Bifidobacterium are well known. The number of established or proposed species has increased only slightly during recent years. 

Only a few species have a long history of use in industrial applications. Bifidobacteria are mainly exploited in dairy products like yogurts or yogurt drinks, but also a whole range of sour milk and other milk based products. Occasionally they are also used in feed in combination with other genera. In Europe only a few species are used (B. animalis, B. longum, B. breve, B. bifidum and B. adolescentis,) and often applied in combination with lactic acid bacteria (Reuter 1990; Reuter 1997; Klein, Pack et al. 1998; Reuter 2002). 

The genome sequences of B. longum (Schell, Karmirantzou et al. 2002) and B. breve have been determined, while the genome sequencing project of B. adolescentis is ongoing. 

1.3 Are there safety concerns? 

Humans. Safety concerns are so far related mainly only to one species, B. dentium, which has been associated with dental caries. It has also been isolated from a case of peritonsillar abscess together with other anaerobes (Civen, Vaisanen et al. 1993) and, under its previous designation “Actinomyces eriksonii”, from pulmonary and subcutaneous abscesses (Slack 1974). Occasionally, other species have been reported to be isolated from human clinical cases, but none of them was the primary cause of disease. Only immunocompromised hosts were infected (Crociani, Biavati et al. 1996). These species are not used as food or feed supplements. None of the bifidobacteria used for industrial purposes have been associated with human clinical disease. 

Although there are few studies on the antibiotic resistance of bifidobacteria strains, the presence of the acquired tetracycline resistance gene tet(W) has been reported in Bifidobacterium animalis subsp. lactis and Bifidobacterium bifidum (Kastner, Perreten et al. 2006; Masco, Van Hoorde et al. 2006). 

Livestock. No report can be found on safety concerns related to Bifidobacteria in animals. 

1.4 Can the safety concerns be excluded? 

There are apparently no specific safety concerns regarding the genus Bifidobacterium (especially concerning B. animalis; B. longum, B. breve, B. adolescentis, and B. bifidum) with the exception of the species associated with dental caries, B. dentium. Susceptibility to antibiotics should be assessed as defined by the EFSA opinion (EFSA 2005) for each strain. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 2 

1.5 Units proposed for QPS status 

Due to the long history of safe use of B. adolescentis, B. animalis; B. longum, B. breve and B. bifidum, these species are proposed for QPS status. Other species could be included subsequent to their industrial application with the exception of the species associated with dental caries (B. dentium). 

2 Corynebacterium 

Corynebacterium glutamicum is a soil bacterium widely used for the biotechnological production of amino acids. Amino acid producing strains have been selected and improved by mutagenesis as well as by using recombinant DNA technology. C. glutamicum belongs to a genus which also includes significant human pathogenic bacteria. Although some Corynebacterium species have been detected as components of the bacterial community of cheese surface, only C. glutamicum is considered of relevance for feed and food sectors. Only this species has been considered for the QPS assessment because of its significant role in the industrial production of amino acids. 

2.1 Taxonomic unit defined 

The genus Corynebacterium belongs to a branch of the Actinomycetales that also includes the genera Mycobacterium, Nocardia and Rhodococcus. Bacterial species belonging to this branch of the Gram-positive bacteria share particular characteristics, such as high G+C content (47– 74%) and a specific cell envelope organisation, mainly characterised by the presence of peptidoglycan, arabinogalactan and mycolic acids. The genus currently contains 63 species, which colonise different environments. 

2.2 Is the body of knowledge sufficient? 

The characteristics, the physiology and the genetics of C. glutamicum are well known. The genome sequence of this industrial bacterium has been determined (Kalinowski, Bathe et al. 2003), reflecting the considerable biotechnological importance of these organisms. 

2.3 Are there safety concerns? 

Corynebacterium glutamicum plays an important role in the amino acid fermentation industry. No safety concerns are reported for this bacterial species for humans and animals, and no information on the presence of acquired antibiotic resistances in this bacterial species is available. However, it should be kept in mind that the direct exposure of consumers to this bacterial species is expected to be very low. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 3 

2.4 Can the safety concerns be excluded? 

1. glutamicum has generally been considered to be non-pathogenic and no safety concerns are envisaged. However, its history of use is as a source of amino acids and has not, to date, involved the direct and deliberate exposure of humans or livestock. 

2.5 Units proposed for QPS status 

There is a long history of safe use of C. glutamicum as an amino acid producer; consequently, C. glutamicum is proposed for QPS status with the qualification that this status applies only when the species is used for production purposes only. 

3 Enterococcus 

Enterococci are significant strains found naturally in some foods and food products. They are often part of the natural microbiota involved in flavour and texture development resulting from fermentation, but can also occur as contaminants of foods. These organisms are used as starter cultures in food products, such as cheese, as probiotic cultures for humans and animals and as silage additives (Franz, Stiles et al. 2003; Foulquie Moreno, Sarantinopoulos et al. 2006). These organisms exist as normal human commensals but they are also associated with human infections. The Enterococcus genus is of particular medical relevance because of its increased incidence as a cause of disease in hospital-acquired (nosocomial) infections and because the available antibiotic therapies are being compromised by evolving transmissible antibiotic resistance. Enterococci harbour virulence factors on mobile genetic elements and they are noted for their capacity to exchange genetic information by bacterial mating. Gene transfer processes are well characterised in Enterococcus (Kozlowicz, Dworkin et al. 2006). 

3.1 Taxonomic unit defined 

In recent years, the genus Enterococcus has undergone considerable changes in taxonomy. Since the recognition of Enterococcus as a separate genus (Schleifer 1984), several new species have been described as a result of improvements in the methods for their identification, combined with a growing interest in their role as opportunistic pathogens. The genus consists currently of the following species: E. aquimarinus, E. asini, E. avium, E. canintestini, E. canis, E. casseliflavus, E. cecorum, E. columbae, E. devriesei, E. dispar, E. durans, E. faecalis, E. faecium, E. gallinarum, E. gilvus, E. haemoperoxidus, E. hermanniensis, E. hirae, E. italicus, E. malodoratus, E. moraviensis, E. mundtii , E. pallens, E. phoeniculicola, E. pseudoavium, E. raffinosus, E. ratti, E. saccharolyticus, E. silesiacus, E. sulfureus and E. termitis. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 4 

3.2 Is the body of knowledge sufficient? 

The characteristics and habitat of most of enterococci species are well known. Strains of E. faecium have a long history of apparent safe use in industrial and agricultural applications. However, members of the same species and same genus are a major concern in clinical practice for their prevalence in nosocomial infections (Zirakzadeh and Patel 2006). The biology, physiology and the genetics of E. faecalis are well known. A number of virulence factors and antibiotic resistance determinants have been identified and characterised. Increased information on this genus is being derived from analysis of the sequence of E. faecalis (Paulsen, Banerjei et al. 2003) and E. faecium (http://genome.ornl.gov/microbial/efae/) genomes (Lepage, Brinster et al. 2006). Although the information on enterococci as infectious agents is available there is still a lack of knowledge on the role of the food chains as a source of virulent enterococci. 

3.3 Are there safety concerns? 

Humans. Enterococci are among the leading causes of both community and hospital-acquired infections. Different virulence factors, implicated in the pathogenesis of enterococci, have been described. In E. faecalis virulence determinants, such as citolysin operon, esp, and the gene encoding aggregation substance are clustered on a large pathogenicity island, a genetic element of approximately 150 kilobases in size (Shankar, Baghdayan et al. 2002; Shankar, Coburn et al. 2004). Cytolysin has been demonstrated to contribute to bacterial virulence (Coburn and Gilmore 2003). Surface adhesin coded on pheromone plasmids, was shown to be a virulence factor involved in the adhesion of enterococci to eukaryotic cell surfaces and in the production of experimental infections (Chandler, Hirt et al. 2005). Gelatinase is an additional virulence factor; E. faecalis mutants with an insertion disruption in the gelE or fsr operon, the gelatinase regulatory system, showed significant delays in mortality in a mouse peritonitis model (Singh, Nallapareddy et al. 2005). Moreover, it has been shown that in E. faecalis gelatinase is important for the translocation across human enterocyte-like T84 cells (Zeng, Teng et al. 2005). Virulence determinants have also been detected in E. faecium. 

The safety concerns related to these bacteria are heightened by the antibiotic resistance of enterococci: these organisms show intrinsic resistance to a variety of antibiotics, such as aminoglycosides and β-lactams. Furthermore, enterococci have acquired genetic determinants for antibiotic resistances, and among them the resistance to high levels of glycopeptides is of major concern (Courvalin 2006). 

Food strains of Enterococcus were found to harbour antibiotic resistance genes (Rizzotti, Simeoni et al. 2005; Hummel, Holzapfel et al. 2007), and high-frequency gene transfer of antibiotic resistance and virulence determinants in food within the enterococcal community in the absence of selective pressure was observed in food (Cocconcelli, Cattivelli et al. 2003). Occurrence of virulence factors in food enterococci is reported (Semedo, Almeida Santos et al. 2003), although 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 5 

they have fewer virulence determinants than did clinical strains (Eaton and Gasson 2001; Mannu, Paba et al. 2003; Lepage, Brinster et al. 2006). 

Livestock. Species of genus Enterococcus, such as E. durans and E. hirae, have been associated with infections in chickens (Chadfield, Christensen et al. 2005; Abe, Nakamura et al. 2006). 

3.4 Can the safety concerns be excluded? 

Some strains of E. faecium show a long history of apparent safe use in food or feed and lack many of the virulence determinants described to date. However, since there are significant safety issues regarding Enterococcus strains, and since the determinants of virulence are not fully understood, concerns cannot be excluded. Considering the number of infections linked to Enterococcus, a case-by-case approach should be adopted. 

3.5 Units proposed for QPS status 

Due to safety concerns and the lack of information on the safety, no members of the genus Enterococcus can be proposed for QPS status. 

4 Lactobacillus 

The genus Lactobacillus is a wide and heterogeneous taxonomic unit, comprising the rod-shaped lactic acid bacteria. This genus encompasses more than 100 different species with a large variety of phenotypic, biochemical and physiological properties. Many of the species are significant constituents of the normal gut flora of humans and livestock although their occurrence and numbers are host dependent. Several species of the genus are intentionally introduced in the food chains, being involved in a range of food and feed fermentations and applied as probiotics for humans and animals. 

4.1 Taxonomic unit defined 

As for other lactic acid bacteria, lactobacilli belong to the phylum Firmicutes. They are rod-shaped, non-motile and non-sporeformers. Classically, the Lactobacillus genus is divided into three groups: group 1, obligate homofermentative, group 2, facultative heterofermentrative and group 3 obligate heterofermentrative (for a review, see Axelsson 2004). The application of phylogenetic molecular taxonomy and 16S rRNA gene sequence analysis resulted in several changes within the taxonomy of this genus, with an increase in the number of species. At present 112 species belong to the genus Lactobacillus. Several molecular methods are available for the identification of lactobacilli to species level. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 6 

4.2 Is the body of knowledge sufficient? 

The characteristics and habitat of most of Lactobacillus species are well known. Some of the species of this genus have a long history of apparent safe use in industrial and agricultural applications. Lactobacilli are used as starter cultures in a variety of food fermentation, such as dairy products, fermented and cured meats, fermented vegetables, sourdough and silage. Moreover, they are among the dominant populations in microbial communities of traditional fermented foods, being part of the natural starter cultures. Increased information on this genus is being derived from the sequence analysis of several genomes of Lactobacillus species. 

4.3 Are there safety concerns? 

Members of the Lactobacillus genus are daily consumed in large quantities in a variety of fermented foods by people of all ages, ethnic groups and health status with apparently no ill effects. Apart from their possible involvement in the development of dental caries, lactobacilli have generally been considered to be non-pathogenic. However, there has been an increasing number of reports that these organisms might occasionally be involved in human disease (Sharpe, Hill et al. 1973; Gasser 1994; Salminen, Rautelin et al. 2006). A variety of different Lactobacillus species has been recovered from human clinical specimens. These include L. rhamnosus, L. fermentum, L. plantarum, L. casei, L. jensenii, L. salivarius, L. gasseri, L. salivarius, and L. acidophilus. Clinical conditions from which these species were derived were chiefly subacute endocarditis and bacteremia or systemic septicemia, but also included abscesses, chorioamnionitis, and urosepsis (Lorenz, Appelbaum et al. 1982; Dickgiesser, Weiss et al. 1984; Salminen, Tynkkynen et al. 2002; Salminen, Rautelin et al. 2004; Salminen, Rautelin et al. 2006). Even the strain L. rhamnosus ATCC 53103, used as human probiotic, has occasionally been encountered in clinical specimens such as blood or pus samples (Rautio, Jousimies-Somer et al. 1999; Salminen, Tynkkynen et al. 2002; Salminen, Rautelin et al. 2004; De Groote, Frank et al. 2005; Salminen, Rautelin et al. 2006). However, Salminen and co-workers (Salminen, Rautelin et al. 2006) demonstrated that increased probiotic use of L. rhamnosus ATCC 53103 had not led to an increase in Lactobacillus bacteraemia. Furthermore, it has been demonstrated that strains isolated from clinical samples, show phenotypic, differences from probiotic L. rhamnosus strains (Klein, Hack et al. 1995; Ouwehand, Saxelin et al. 2004). Many of the patients with apparent Lactobacillus infection were immunocompromised or had other severe underlying illnesses. As far as endocarditis due to lactobacilli is concerned, this infection usually develops on the basis of preceding anatomical alterations of the heart valves. There are indications, however, that good adhesion properties of lactobacilli and, thus, of probiotic strains, might be a potential risk for bacteremia (Apostolou, Kirjavainen et al. 2001). In conclusion, most of the Lactobacillus species described to date can rightly be considered to be non-pathogenic to humans (Bernardeau, Guguen et al. 2006). Only certain strains of L. rhamnosus may be considered to be potential human opportunistic pathogens because they not only affect severely immunocompromised, but also 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 7 

immunologically healthy individuals with a history of rheumatic endocarditis or heart valve replacement. 

Several examples of antibiotic resistant lactobacilli isolated from food or from the gut of animals exist. Acquired genes for antibiotic resistance have been detected in Lactobacillus species: tet(M) has been found in L. plantarum, L. brevis, L. sakei and L. curvatus (Danielsen 2002; Gevers, Danielsen et al. 2003) and tet(S) in L. plantarum (Huys, D'Haene et al. 2006). Erythromycin resistance determinants erm(B) has been found in L. plantarum, L. salivarius, L. animalis, L. fermentum, L. reuteri (Axelsson, Ahrne et al. 1988; Fons, Hege et al. 1997; Gevers, Danielsen et al. 2003; Martel, Meulenaere et al. 2003). Moreover, the gene coding for the bifunctional aminoglycoside-modifying enzyme AAC(6')-APH(2") was detected in L. salivarius and L. acidophilus (Tenorio, Zarazaga et al. 2001) and chloramphenicol resistance gene cat was identified in L. reuteri (Lin, Fung et al. 1996). Obligate and facultative heterofermentative lactobacilli, and L. salivarius, are intrinsically resistant to vancomycin and other glycopeptide antibiotics. 

Several genetic determinants for antibiotic resistance in Lactobacillus are harboured by extrachromosomal elements (Lin, Fung et al. 1996; Danielsen 2002; Gevers, Danielsen et al. 2003; Gfeller, Roth et al. 2003; Huys, D'Haene et al. 2006). However, transferable elements encoding resistances of clinical relevance, such as to the glycopetides have been excluded for some probiotic L. reuteri and L. rhamnosus strains (Klein, Hallmann et al. 2000). 

Livestock. No report can be found on safety concerns related to lactobacilli in animals 

4.4 Can the safety concerns be excluded? 

There are apparently no specific safety concerns regarding a number of Lactobacillus species which have a long history of apparent safe use in the food chain. Susceptibility to antibiotics should be assessed as defined by the EFSA opinion for each strain (EFSA 2005). 

4.5 Units proposed for QPS status 

Due to the long history of safe use the following species are proposed for QPS status: 

1. acidophilus, L. amylolyticus, L. amylovorus, L. alimentarius, L. aviaries, L. brevis, L. buchneri, L. casei, L. crispatus, L. curvatus, L. delbrueckii, L. farciminis, L. fermentum, L. gallinarum, L. gasseri, L. helveticus, L. hilgardii, L. johnsonii, L. kefiranofaciens, L. kefiri, L. mucosae, L. panis, L. paracasei, L. paraplantarum, L. pentosus, L. plantarum, L. pontis, L. reuteri, L. rhamnosus, L. sakei, L. salivarius, L. sanfranciscensis and L. zeae. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 8 

5 Lactococcus 

The genus Lactococcus was previously known as group N-streptococci. The type species, Lactococcus lactis, (formerly Streptococcus lactis) is a well known dairy starter organism and members of this species are common component of cheese bacterial communities. The other recognised species do not have a history of intentional food use. 

5.1 Taxonomic unit defined 

As for other lactic acid bacteria, lactococci belong to the phylum Firmicutes. They are coccoid, non-motile organism metabolising hexoses homofermentatively and pentoses heterofermentatively. They are mesophilic with temperature optima generally at 30oC or less (for a review, see Axelsson 2004). The genus consists of five species: Lactococcus lactis (with subspecies L. lactis subsp. lactis, L. lactis subsp. cremoris and L. lactis subsp. hordniae), L. garvieae, L. plantarum, L. raffinolactis, and L. piscium (Schleifer 1987; Williams, Fryer et al. 1990). 

5.2 Is the body of knowledge sufficient? 

The characteristics and habitat of the dairy species of L. lactis are well known, and they are extensively used as starters for the production of cheese and fermented milks. Complete genomic sequences of several industrial and laboratory strains are available (see Morelli 2004 for review). Outside the dairy environment natural habitats of lactococci include plant material (Kelly, Davey et al. 2000) and fish (Ringǿ 2004). L. garviae and L. piscium are well-known as fish pathogens (Williams, Fryer et al. 1990; Eyngor, Zlotkin et al. 2004). 

5.3 Are there safety concerns? 

Humans. Lactococcus lactis is consumed in large quantities in cheese and fermented milks by people of all ages, ethnic groups and variable health status with apparently no ill effects. Indeed, the relatively low growth temperature optima make even opportunistic infections unlikely. However, rare cases of endocarditis (Wood, Jacobs et al. 1955; Mannion and Rothburn 1990; Pellizzer, Benedetti et al. 1996; Halldorsdottir, Haraldsdottir et al. 2002) have been reported as well as septicemia (Durand, Rousseau et al. 1995) necrotising pneumonitis (Torre, Sampietro et al. 1990), septic arthritis (Campbell, Dealler et al. 1993), cerebral abscess (Akhaddar, El Mostarchid et al. 2002) and liver abscess (Nakarai, Morita et al. 2000). In the majority of these cases there have been predisposing factors, such as underlying disease, immunocompromised status or early age, although liver abscess caused by L. lactis in an immunocompetent adult has recently been reported (Antolin, Ciguenza et al. 2004). However, these infections represent extremely rare individual cases, and should not be regarded as an indication of human pathogenicity taking into account the extent of exposure to these microorganisms. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 9 

In general, dairy lactococci are sensitive to most clinical antibiotics to a degree that antibiotic residues in milk can cause starter failures. According to the results of (de Fabrizio 1994) strains of L. lactis were sensitive to ampicillin and other β-lactams (oxacillin, penicillin, pipericillin, and certain cephalosporins), chloramphenicol, erythromycin, amikacin, gentamicin, tetracycline, sulphonamide, trimethoprim/sulfamethoxazole and vancomycin. Somewhat lowered susceptibility towards carbenicillin, ciprofloxacin, dicloxacillin and norfloxacin, and intrinsic resistance towards colistin, fosfomycin, pipedimic acid and rifamycin were observed. 

Lactococci are well known to contain plasmids and to exchange genetic material by intra- and intergeneric conjugation (see Morelli 2004 for a review), and the potential for the spread of transferable antibiotic resistances thus exists. Indeed, pK214, a plasmid from a L. lactis strain originally isolated from a raw milk soft cheese, harbours resistance determinants for streptomycin (streptomycin adenylase), tetracycline (Tet S) and chloramphenicol (chloramphenicol acetylase) and an efflux protein conferring resistance to macrolides in Escherichia coli, but not in the strain itself (Perreten, Schwarz et al. 1997). Since all these resistance determinants showed homologies to genes resident in other species (Streptococcus pyogenes, Staphylococcus aureus and Listeria monocytogenes) pK24 illustrates the potential for genetic exchange even between taxonomically distant species. 

Livestock. Lactococcus garviae and L. piscium, are well known fish pathogens (Williams, Fryer et al. 1990; Torranzo 2005), the former dominating in the warm water species and the latter at temperatures below 15oC. Moreover, L. garviae was originally isolated from the udder of a mastitic cow (Collins, Farrow et al. 1983). Thus these species have a distinct pathogenic potential in aquaculture, and maybe also in other livestock. 

5.4 Can the safety concerns be excluded? 

With the dairy lactococci the possibility of human or veterinary infections, at least in warm blooded animals, are extremely remote, and in practice this is not a concern. Thus with the dairy strains the main safety issue is the presence of acquired antibiotic resistances, which should be monitored. 

With the other lactococcal species there is a lack of history of intentional food use, and the pathogenicity of certain species (L. garviae, L. piscium) in fish necessitates strain-specific safety assessment in eventual applications related to aquaculture. 

5.5 Units proposed for QPS status 

The dairy species of L. lactis (L. lactis subsp. lactis, its biovariant diacetylactis, and L. lactis subsp. cremoris) have a long history as dairy starters and an excellent safety record. The occasional and extremely rare infections, in which also these organisms have been associated, do 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 10 

not warrant specific safety concerns. Thus these subspecies can be proposed for QPS status, provided that the presence of acquired antibiotic resistance determinants has been excluded (EFSA 2005). 

6 Leuconostoc 

The Leuconostoc genus contains obligate heterofermentrative lactic acid cocci. These bacteria have as their predominant natural habitat plants and fermented food of plant origin. Due to their positive contribution, members of this genus are frequently used as a starter cultures to promote food fermentation and malolactic fermentation in wine. Moreover, Leuconostoc species play a significant role in dairy fermentation, where they contribute to the aroma formation, and also in meat fermentation. 

6.1 Taxonomic unit defined 

Bacteria of the genus Leuconostoc belong to phylum Firmicutes. They are coccoid, non-motile organisms, showing obligate heterofermentative metabolism. They are mesophilic with temperature optima generally at 30oC or less. The genus consists currently of the following species: L. carnosum, L. citreum, L. durionis, L. fallax, L. ficulneum, L. fructosum,, L. gelidum, L. inhae, L. kimchii, L. lactis, L. mesenteroides subsp cremoris, L. mesenteroides subsp dextranicums, L. mesenteroides subsp mesenteroides, Leuconostoc pseudoficulneum L. pseudomesenteroide. 

6.2 Is the body of knowledge sufficient? 

The characteristics and habitat of the Leuconostoc are well known, and some species have a long history of apparent safe use as starter cultures for dairy, wine and vegetable fermentations. Moreover, they are among the dominant populations in microbial communities of several traditional fermented foods. Projects to determine the nucleotide sequence of the genome sequencing projects of L. mesenteroides are ongoing. 

6.3 Are there safety concerns ? 

Humans. Leuconostoc have generally been considered to be non-pathogenic bacteria. However, there have been reports that these organisms might occasionally be involved in human disease (Vagiakou-Voudris, Mylona-Petropoulou et al. 2002; Kumudhan and Mars 2004). Infections by Leuconostoc species are uncommon, and usually affect patients with an underlying disease, therefore Leuconostoc species described so far can rightly be considered to be non-pathogenic to humans. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 11 

Although there are few studies on the antibiotic resistance of Leuconostoc, strains from food isolates were also found carrying antibiotic resistance genes (Wang, Manuzon et al. 2006). 

Livestock. No report can be found on safety concerns related to Leuconostoc species in animals. 

6.4 Can the safety concerns be excluded? 

There apparently are no specific safety concerns regarding L. citreum, L. lactis, and L. mesenteroides, which have a long history of apparent safe use in the food chain. Susceptibility to antibiotics should be assessed as defined by EFSA for each strain (EFSA 2005). 

6.5 Units proposed for QPS status 

Due to the long history of safe use, L. citreum, L. lactis, and L. mesenteroides species are proposed for QPS status. 

7 Pediococcus 

Pediococci are Gram-positive, coccus-shaped, lactic acid bacteria, showing the distinctive characteristic of formation of tetrads of cells, via cell division in two perpendicular directions in a single plane. Pediococci have been isolated from a variety of food fermentations, such as cheese, sausages and fermented plant products. Pediococci are also involved in spoilage of wine, beer and other alcoholic beverages. Due to their positive role in food fermentation, members of this genus are used as starter cultures for dairy, meat and vegetable fermentations. 

7.1 Taxonomic unit defined 

As for other lactic acid bacteria, pediococci belong to phylum Firmicutes. Phylogenetically all species of Pediococcus fall within the Lactobacillus cluster of lactic acid and related Gram-positive bacteria and forms a Lactobacillus casei – Pediococcus sub-cluster. The genus consists currently of the following nine species: P. acidilactici, P. cellicola, P. claussenii, P. damnosus, P. dextrinicus, P. inopinatus, P. parvulus, P. pentosaceus and P. stilesii. 

7.2 Is the body of knowledge sufficient? 

The characteristics and habitat of the species of the Pediococcus genus are well known, and some species have a long history of apparent safe use as starter cultures for dairy, sausage and vegetable fermentations. Moreover, they are among the dominant populations in microbial communities of several traditional fermented foods. Projects to determine the nucleotide sequence of P. pentosaceus are ongoing (http://genome.jgi-psf.org/draft_microbes/pedpe/pedpe.home.html). 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 12 

7.3 Are there safety concerns? 

Pediococci are consumed in large quantities in cheese and fermented sausages by people of all ages, ethnic groups and health status with apparently no ill-effects. Pediococci have generally been considered to be non-pathogenic. Pediococcus spp. are rarely isolated from clinical specimens, and there are few reports on their involvement in human disease (Heinz, von Wintzingerode et al. 2000), usually only affecting patients with an underlying disease. Therefore pediococci can be considered to be non-pathogenic to humans. 

Acquired genes for antibiotic resistance have been detected in Pediococcus genus: tet(M) and tet(S) genetic determinants for tetracycline resistance have been found in P. pentosaceus (Gevers, Danielsen et al. 2003). Tankovic, Leclercq et al. (1993) reported the presence of erythromycin resistance determinants, homologous to ermAM, carried by a 46-kb plasmid, pVM20 in P. acidilactici. Moreover, the gene coding for the bifunctional aminoglycoside-modifying enzyme AAC(6')-APH(2") was detected in Pedioccoccus acidilactici (Tenorio, Zarazaga et al. 2001). 

7.4 Can the safety concerns be excluded? 

There are apparently no specific safety concerns regarding P. acidilactici, P. dextrinicus and P. pentosaceus, which have a long history of apparent safe use in the food chain. Susceptibility to antibiotics should be assessed as defined by EFSA for each strain (EFSA 2005). 

7.5 Units proposed for QPS status 

Due to the long history of safe use, the species P. acidilactici, P. dextrinicus and P. pentosaceus are recommended for QPS status. 

8 Propionibacteria 

Propionic acid bacteria have a long history of use as aroma producers in dairy products, and they have also been used as silage inoculants, due to the antifungal properties of propionic acid. The genus can be divided into dairy species and species associated with human or animal skin. 

8.1 Taxonomic unit defined 

Propionic acid bacteria (PAB) belong to the Actinomyctes-branch of Firmicutes.. They are mesophilic, non-sporing, non-motile irregular rods. The metabolism is anaerobic and the main fermentation endproducts are propionic acid, acetic acid and CO2. PAB can be divided into dairy species (DPAB) and mucocutaneous PAB. The former include: P. acidopropionici, P. australiense, P. cyclohexanicum, P. freudenreichii subsp. freudenreichii, P. freudenreichii 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 13 

subsp. shermanii, P. jensenii, P. thoenii and P. microaerophilum. The cutaneous species are P. acnes, P. avidum, P. granulosum, P. lymohophilum and P. propionicum (For a review, see Ouwehand 2004). 

8.2 Is the body of knowledge sufficient? 

DPAB have been traditionally associated with certain types of cheese (Swiss cheese or Emmenthal being the best known), although they have been occasionally isolated from rumen and as contaminants from spoiled foodstuffs. The mucocutaneous PAB belong to the normal microbiota of human skin and/or mucous membranes, and can occasionally be isolated from the faeces. 

The traditional use of DPAB is as cheese starters. Their main function is to produce propionic acid, which is an important aroma component and to form the characteristic “eyes” in cheese through development of CO2. Propionic acid has also antimicrobial properties, particularly against fungi. In addition PAB are known to produce a variety of bacteriocins (Ouwehand 2004), such as acnecin, jenseniins P and G, propionicins PLG-1, T1 SM1 and SM2, and protease-activated antimicrobial peptide (PAMP). The producing species are mainly cutaneous strains, although strains of P. freudenreichii subsp. shermanii are also known to produce acnecin. The antimicrobial spectrum of these bacteriocins cover other PAB, lactic acid bacteria, other Gram-positive and Gram-negative bacteria as well as yeasts and filamentous fungi in some cases. 

Because of their antimicrobial action DPAB are also used as components of protective cultures that are used to prevent microbial spoilage of foodstuffs (Ayres 1992; Mäyrä-Mäkinen 1995). The use of PAB as silage starters is also mainly based on their antifungal action. 

8.3 Are there safety concerns? 

Humans: Although mucocutaneous PAB are normal commensals on human skin or mucous membranes, they, especially P. acnes and P. propionicum, have been also associated with the pathogenesis of acne (Bojar and Holland 2004) and various other infections (Al-Mazrou 2005; Lutz, Berthelot et al. 2005). P. propionicum, possibly better known among infectious disease under its previous designations “Actinomyces propionicus”or “Arachnia propionica”, is one of the less frequent, but nonetheless typical causes of human actinomycosis, a disease which has been known for decades for its notorious diagnostic and therapeutic problems. Furthermore, P. propionicum is the most characteristic and frequent causative agent of human lachrymal canaliculitis with and without conjunctivitis (Brazier and Hall 1993). 

The complete genomic sequence of the organism P. acnes is available, and might eventually provide the determinants of pathogenicity allowing these species to be differentiated from DPAB (Bruggemann 2005). 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 14 

Due to their use as cheese starters, relatively large amounts of DPAB are consumed by humans in Europe without any observable ill effects. Toxicological studies on P. freudenreichii subsp. shermanii have been reported. Intraperitoneal doses of 109 – 1010 cfu did not cause observable ill effects in mice, guinea pigs or rabbits, nor did it have any cytotoxic effect on tissue cultures (Sidorchuk and Bondarenko 1984). 

The information on antibiotic resistance of Propionibacteria is limited to mucocutanoeous species, in particular to P. acnes, P. granulosum, and P. avidum and P. propionicum, and to antibiotics used either topically or systematically to combat infections. Tetracycline and erythromycin resistances have been mainly associated with base change mutations in 16S rDNA. A transposon- associated erythromycin resistance determinant conferring resistance to MLS antibiotics has also been characterised among the clinical isolates (Ross, Snelling et al. 2001; Ross, Snelling et al. 2003). The presence of acquired determinants for antimicrobial resistance has not been reported in P. freudenreichii. 

Livestock: No reports of veterinary problems associated with the use of DPAB in silage could be found. However, there is an Australian report of four cases of abscess-like gross lesions in carcases of cattle, where the associated microorganism has been characterised as closely related to P. cyclohexanicum/P.freudenreichii cluster (Forbes-Faulkner, Pitt et al. 2000). 

8.4 Can the safety concerns be excluded? 

There is a long history of apparent safe use of DPAB, particularly P. freudenreichii subsp. freudenreichii and P. freudenreichii subsp. shermanii in dairy foods and to certain extent also in silage. Although a closer inspection might reveal occasional cases of opportunistic infections associated with microorganisms closely related to these bacteria, these do not warrant specific safety evaluation of taxonomically well characterised P. freudenreichii and its subspecies or closely related dairy species. 

Since the data from mucocutaneous strains show that acquired antibiotic resistance can be associated with Propionibacteria, susceptibility to antibiotics should be assessed as defined by EFSA for each strain (EFSA 2005). 

8.5 Units proposed for QPS status 

There is a long history of safe use of DPAB, particularly P. freudenreichii subsp. freudenreichii and P. freudenreichii subsp. shermanii in dairy foods and to certain extent also in silage. Occasional cases of opportunistic infections associated with microorganisms closely related to these bacteria may be found but these do not warrant specific safety evaluation of taxonomically well characterised P. freudenreichii and its subspecies. Consequently, P. freudenreichii is proposed for QPS status, provided that the absence of acquired antibiotic resistances is demonstrated as indicated in the EFSA opinion (EFSA 2005). 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 15 

9 Streptococcus thermophilus 

1. thermophilus is a relevant dairy starter microorganism, used for the manufacture of fermented milks and cheese. This species belongs to the genus Streptococcus, a taxonomical unit which includes pathogenic and oral streptococci. Since S. thermophilus is an exception in this genus, for its relevant role as industrial and food organism, only this species has been considered for the QPS assessment. 

9.1 Taxonomic unit defined 

The taxonomy of S. thermophilus has been quite controversial, it has been considered a subspecies of S. salivarius and more recently this species has been revived on the basis of molecular taxonomy (Schleifer 1991). 

9.2 Is the body of knowledge sufficient? 

1. thermophilus is widely used as starter cultures for cheese and fermented milk. The biology of this microorganism is well known and three different genomes sequences are available (Hols, Hancy et al. 2005). 

9.3 Are there safety concerns? 

1. thermophilus is consumed in large quantities in cheese and fermented milks by people of all ages, ethnic groups and variable health status with apparently no ill effects. This species has generally been considered to be non-pathogenic and no safety concerns are envisaged. 

Although there are few reports on the antibiotic resistance/susceptibility in S. thermophilus, acquired resistance genes has been detected in this species. Thus, genetic determinants for tetracycline tet(S) and erythromycin erm(B) resistance were detected in dairy strains of S. thermophilus (Wang, Manuzon et al. 2006). 

9.4 Can the safety concerns be excluded/Qualifications? 

Since there are not safety concerns the only qualification required is the assessment of susceptibility to antibiotics for each single strain, as defined by EFSA. 

9.5 Units proposed for QPS status 

There are no safety concerns for S. thermophilus, therefore QPS status is proposed for this species, provided that the lack of acquired antibiotic resistance is demonstrated (EFSA 2005). 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 16 

REFERENCES 

Abe, Y., K. Nakamura, et al. (2006). "Encephalomalacia with Enterococcus durans infection in the brain stem and cerebral hemisphere in chicks in Japan." Avian Dis 50(1): 139-41. 

Akhaddar, A., B. El Mostarchid, et al. (2002). "Cerebellar abscess due to Lactococcus lactis. A new pathogen." Acta Neurochir (Wien) 144(3): 305-6. 

Al-Mazrou, A. M. (2005). "Propionibacterium acnes. A cause of pneumatocele associated pneumonia." Saudi Med J 26(7): 1127-9. 

Antolin, J., R. Ciguenza, et al. (2004). "Liver abscess caused by Lactococcus lactis cremoris: a new pathogen." Scand J Infect Dis 36(6-7): 490-1. 

Apostolou, E., P. V. Kirjavainen, et al. (2001). "Good adhesion properties of probiotics: a potential risk for bacteremia?" FEMS Immunol Med Microbiol 31(1): 35-9. 

Axelsson, L. T. (2004). Lactic acid bacteria: Classification and physiology. Lactic Acid Bacteria. Microbiological and Functional Aspects. S. Salminen, Ouwehand, A., von Wright, A. New York, Marcel Dekker Inc.: 1-66. 

Axelsson, L. T., S. E. Ahrne, et al. (1988). "Identification and cloning of a plasmid-encoded erythromycin resistance determinant from Lactobacillus reuteri." Plasmid 20(2): 171-4. 

Ayres, J. W., Sandine, W.E. and Weber, G.H. (1992). Preserving foods using metabolites of propionibacteria other than propionic acid. US Patent. 5 096 718. 

Bernardeau, M., M. Guguen, et al. (2006). "Beneficial lactobacilli in food and feed: long-term use, biodiversity and proposals for specific and realistic safety assessments." FEMS Microbiol Rev 30(4): 487-513. 

Bojar, R. A. and K. T. Holland (2004). "Acne and Propionibacterium acnes." Clin Dermatol 22(5): 375-9. 

Brazier, J. S. and V. Hall (1993). "Propionibacterium propionicum and infections of the lacrimal apparatus." Clin Infect Dis 17(5): 892-3. 

Bruggemann, H. (2005). "Insights in the pathogenic potential of Propionibacterium acnes from its complete genome." Semin Cutan Med Surg 24(2): 67-72. 

Campbell, P., S. Dealler, et al. (1993). "Septic arthritis and unpasteurised milk." J Clin Pathol 46(11): 1057-8. 

Chadfield, M. S., J. P. Christensen, et al. (2005). "Characterization of Enterococcus hirae outbreaks in broiler flocks demonstrating increased mortality because of septicemia and endocarditis and/or altered production parameters." Avian Dis 49(1): 16-23. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 17 

Chandler, J. R., H. Hirt, et al. (2005). "A paracrine peptide sex pheromone also acts as an autocrine signal to induce plasmid transfer and virulence factor expression in vivo." Proc Natl Acad Sci U S A 102(43): 15617-22. 

Civen, R., M. L. Vaisanen, et al. (1993). "Peritonsillar abscess, retropharyngeal abscess, mediastinitis, and nonclostridial anaerobic myonecrosis: a case report." Clin Infect Dis 16 Suppl 4: S299-303. 

Coburn, P. S. and M. S. Gilmore (2003). "The Enterococcus faecalis cytolysin: a novel toxin active against eukaryotic and prokaryotic cells." Cell Microbiol 5(10): 661-9. 

Cocconcelli, P. S., D. Cattivelli, et al. (2003). "Gene transfer of vancomycin and tetracycline resistances among Enterococcus faecalis during cheese and sausage fermentations." Int J Food Microbiol 88(2-3): 315-23. 

Collins, M. D., J. A. Farrow, et al. (1983). "Streptococcus garvieae sp. nov. and Streptococcus plantarum sp. nov." J Gen Microbiol 129(11): 3427-31. 

Courvalin, P. (2006). "Vancomycin resistance in gram-positive cocci." Clin Infect Dis 42 Suppl 1: S25-34. 

Crociani, F., B. Biavati, et al. (1996). "Bifidobacterium inopinatum sp. nov. and Bifidobacterium denticolens sp. nov., two new species isolated from human dental caries." Int J Syst Bacteriol 46(2): 564-71. 

Danielsen, M. (2002). "Characterization of the tetracycline resistance plasmid pMD5057 from Lactobacillus plantarum 5057 reveals a composite structure." Plasmid 48(2): 98-103. 

de Fabrizio, S., Parada, JL., and Ledford, RA. (1994). "Antibiotic resistance of Lactococcus lactis - an approach of genetic determinants location through a model system." Microb. Aliment. Nutr. 12: 307-315. 

De Groote, M. A., D. N. Frank, et al. (2005). "Lactobacillus rhamnosus GG bacteremia associated with probiotic use in a child with short gut syndrome." Pediatr Infect Dis J 24(3): 278-80. 

Dickgiesser, U., N. Weiss, et al. (1984). "Lactobacillus gasseri as the cause of septic urinary infection." Infection 12(1): 14-6. 

Durand, J. M., M. C. Rousseau, et al. (1995). "Streptococcus lactis septicemia in a patient with chronic lymphocytic leukemia." Am J Hematol 50(1): 64-5. 

Eaton, T. J. and M. J. Gasson (2001). "Molecular screening of Enterococcus virulence determinants and potential for genetic exchange between food and medical isolates." Appl Environ Microbiol 67(4): 1628-35. 

EFSA (2005). "Opinion of the Scientific Committee on a request from EFSA related to a generic approach to the safety assessment by EFSA of microorganisms used in food/feed and the production of food/feed additives." The EFSA Journal 226: 1-12. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 18 

EFSA (2005). "Opinion of the Scientific Panel on Additives and Products or Substances used in Animal Feed on the updating of the criteria used in the assessment of bacteria for resistance to antibiotics of human or veterinary importance." The EFSA Journal 223: 1-12. 

Eyngor, M., A. Zlotkin, et al. (2004). "Clonality and diversity of the fish pathogen Lactococcus garvieae in Mediterranean countries." Appl Environ Microbiol 70(9): 5132-7. 

Fons, M., T. Hege, et al. (1997). "Isolation and characterization of a plasmid from Lactobacillus fermentum conferring erythromycin resistance." Plasmid 37(3): 199-203. 

Forbes-Faulkner, J. C., D. Pitt, et al. (2000). "Novel Propionibacterium infection in cattle." Aust Vet J 78(3): 175-8. 

Foulquie Moreno, M. R., P. Sarantinopoulos, et al. (2006). "The role and application of enterococci in food and health." Int J Food Microbiol 106(1): 1-24. 

Franz, C. M., M. E. Stiles, et al. (2003). "Enterococci in foods--a conundrum for food safety." Int J Food Microbiol 88(2-3): 105-22. 

Gasser, F. (1994). "Safety of lactic acid bacteria and their occurrence in human clinical infections." Bull. Inst. Pasteur 92: 45-67. 

Gevers, D., M. Danielsen, et al. (2003). "Molecular characterization of tet(M) genes in Lactobacillus isolates from different types of fermented dry sausage." Appl Environ Microbiol 69(2): 1270-5. 

Gfeller, K. Y., M. Roth, et al. (2003). "Sequence and genetic organization of the 19.3-kb erythromycin- and dalfopristin-resistance plasmid pLME300 from Lactobacillus fermentum ROT1." Plasmid 50(3): 190-201. 

Halldorsdottir, H. D., V. Haraldsdottir, et al. (2002). "Endocarditis caused by Lactococcus cremoris." Scand J Infect Dis 34(3): 205-6. 

Heinz, M., F. von Wintzingerode, et al. (2000). "A case of septicemia with Pediococcus acidilactici after long-term antibiotic treatment." Eur J Clin Microbiol Infect Dis 19(12): 946-8. 

Hols, P., F. Hancy, et al. (2005). "New insights in the molecular biology and physiology of Streptococcus thermophilus revealed by comparative genomics." FEMS Microbiol Rev 29(3): 435-63. 

Hummel, A., W. H. Holzapfel, et al. (2007). "Characterisation and transfer of antibiotic resistance genes from enterococci isolated from food." Syst Appl Microbiol 30(1): 1-7. 

Huys, G., K. D'Haene, et al. (2006). "Genetic basis of tetracycline and minocycline resistance in potentially probiotic Lactobacillus plantarum strain CCUG 43738." Antimicrob Agents Chemother 50(4): 1550-1. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 19 

Kalinowski, J., B. Bathe, et al. (2003). "The complete Corynebacterium glutamicum ATCC 13032 genome sequence and its impact on the production of L-aspartate-derived amino acids and vitamins." J Biotechnol 104(1-3): 5-25. 

Kastner, S., V. Perreten, et al. (2006). "Antibiotic susceptibility patterns and resistance genes of starter cultures and probiotic bacteria used in food." Syst Appl Microbiol 29(2): 145-55. 

Kelly, W. J., G. P. Davey, et al. (2000). "Novel sucrose transposons from plant strains of Lactococcus lactis." FEMS Microbiol Lett 190(2): 237-40. 

Klein, G., B. Hack, et al. (1995). "Intra-species characterization of clinical isolates and biotechnologically used strains of Lactobacillus rhamnosus by analysis of the total soluble cytoplasmatic proteins with silver staining." Int J Food Microbiol 25(3): 263-75. 

Klein, G., C. Hallmann, et al. (2000). "Exclusion of vanA, vanB and vanC type glycopeptide resistance in strains of Lactobacillus reuteri and Lactobacillus rhamnosus used as probiotics by polymerase chain reaction and hybridization methods." J Appl Microbiol 89(5): 815-24. 

Klein, G., A. Pack, et al. (1998). "Taxonomy and physiology of probiotic lactic acid bacteria." Int J Food Microbiol 41(2): 103-25. 

Kozlowicz, B. K., M. Dworkin, et al. (2006). "Pheromone-inducible conjugation in Enterococcus faecalis: a model for the evolution of biological complexity?" Int J Med Microbiol 296(2-3): 141- 7. Kumudhan, D. and S. Mars (2004). "Leuconostoc mesenteroids as a cause of post-operative endophthalmitis--a case report." Eye 18(10): 1023-4. 

Lepage, E., S. Brinster, et al. (2006). "Comparative genomic hybridization analysis of Enterococcus faecalis: identification of genes absent from food strains." J Bacteriol 188(19): 6858-68. 

Lin, C. F., Z. F. Fung, et al. (1996). "Molecular characterization of a plasmid-borne (pTC82) chloramphenicol resistance determinant (cat-TC) from Lactobacillus reuteri G4." Plasmid 36(2): 116-24. 

Lorenz, R. P., P. C. Appelbaum, et al. (1982). "Chorioamnionitis and possible neonatal infection associated with Lactobacillus species." J Clin Microbiol 16(3): 558-61. 

Lutz, M. F., P. Berthelot, et al. (2005). "Arthroplastic and osteosynthetic infections due to Propionibacterium acnes: a retrospective study of 52 cases, 1995-2002." Eur J Clin Microbiol Infect Dis 24(11): 739-44. 

Mannion, P. T. and M. M. Rothburn (1990). "Diagnosis of bacterial endocarditis caused by Streptococcus lactis and assisted by immunoblotting of serum antibodies." J Infect 21(3): 317-8. 

Mannu, L., A. Paba, et al. (2003). "Comparison of the incidence of virulence determinants and antibiotic resistance between Enterococcus faecium strains of dairy, animal and clinical origin." Int J Food Microbiol 88(2-3): 291-304. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 20 

Martel, A., V. Meulenaere, et al. (2003). "Macrolide and lincosamide resistance in the gram- positive nasal and tonsillar flora of pigs." Microb Drug Resist 9(3): 293-7. 

Masco, L., K. Van Hoorde, et al. (2006). "Antimicrobial susceptibility of Bifidobacterium strains from humans, animals and probiotic products." J Antimicrob Chemother 58(1): 85-94. 

Mäyrä-Mäkinen, A., Suomalainen, T. (1995). Lactobacillus casei subsp. rhamnosus, bacterial preparations comprising said strain and and use of said strain and preparations for controlling yeasts and moulds. U.S. patent. 5 378 458. 

Morelli, L., Vogensen, F. and von Wright, A. (2004). Genetics of lactic acid bacteria. Lactic Acid Bacteria. Microbiological and Functional Aspects. S. Salminen, Ouwehand, A., von Wright, A. New York, Marcel Dekker Inc.: 249-293. 

Nakarai, T., K. Morita, et al. (2000). "Liver abscess due to Lactococcus lactis cremoris." Pediatr Int 42(6): 699-701. 

Ouwehand, A. (2004). The probiotic potential of Propionibacteria. Lactic Acid Bacteria. Microbiological and Functional Aspects. S. Salminen, Ouwehand, A., von Wright, A. New York, Marcel Dekker, Inc.: 159 - 174. 

Ouwehand, A. C., M. Saxelin, et al. (2004). "Phenotypic differences between commercial Lactobacillus rhamnosus GG and L. rhamnosus strains recovered from blood." Clin Infect Dis 39(12): 1858-60. 

Paulsen, I. T., L. Banerjei, et al. (2003). "Role of mobile DNA in the evolution of vancomycin- resistant Enterococcus faecalis." Science 299(5615): 2071-4. 

Pellizzer, G., P. Benedetti, et al. (1996). "Bacterial endocarditis due to Lactococcus lactis subsp. cremoris: case report." Clin Microbiol Infect 2(3): 230-232. 

Perreten, V., F. Schwarz, et al. (1997). "Antibiotic resistance spread in food." Nature 389(6653): 801-2. 

Rautio, M., H. Jousimies-Somer, et al. (1999). "Liver abscess due to a Lactobacillus rhamnosus strain indistinguishable from L. rhamnosus strain GG." Clin Infect Dis 28(5): 1159-60. 

Reuter, G. (1990). "Bifidobacteria cultures as components of yoghurt-like products." Bifidobacteria Microflora 9: 107-118. 

Reuter, G. (1997). "Present and future of probiotics in Germany and central Europe." Biosc. Microfl. 16: 43-51. 

Reuter, G., Klein, G. and Goldberg, M. (2002). "Identification of probiotic cultures in food samples." Food Research International 35: 117-124. 

Ringǿ, E. (2004). Lactic acid bacteria in fish and in fish farming. Lactic Acid Bacteria. Microbiological and Functional Aspects. S. Salminen, Ouwehand, A., von Wright, A. New York, Marcel Dekker Inc.: 581-610. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 21 

Rizzotti, L., D. Simeoni, et al. (2005). "Contribution of enterococci to the spread of antibiotic resistance in the production chain of swine meat commodities." J Food Prot 68(5): 955-65. 

Ross, J. I., A. M. Snelling, et al. (2003). "Antibiotic-resistant acne: lessons from Europe." Br J Dermatol 148(3): 467-78. 

Ross, J. I., A. M. Snelling, et al. (2001). "Phenotypic and genotypic characterization of antibiotic- resistant Propionibacterium acnes isolated from acne patients attending dermatology clinics in Europe, the U.S.A., Japan and Australia." Br J Dermatol 144(2): 339-46. 

Salminen, M. K., H. Rautelin, et al. (2004). "Lactobacillus bacteremia, clinical significance, and patient outcome, with special focus on probiotic L. rhamnosus GG." Clin Infect Dis 38(1): 62-9. 

Salminen, M. K., H. Rautelin, et al. (2006). "Lactobacillus bacteremia, species identification, and antimicrobial susceptibility of 85 blood isolates." Clin Infect Dis 42(5): e35-44. 

Salminen, M. K., S. Tynkkynen, et al. (2002). "Lactobacillus bacteremia during a rapid increase in probiotic use of Lactobacillus rhamnosus GG in Finland." Clin Infect Dis 35(10): 1155-60. 

Schell, M. A., M. Karmirantzou, et al. (2002). "The genome sequence of Bifidobacterium longum reflects its adaptation to the human gastrointestinal tract." Proc Natl Acad Sci U S A 99(22): 14422-7. 

Schleifer, K. H., and R. Kilpper-Balz (1987). "Molecular and chemotaxonomic approaches to the classification of streptococci, enterococci and lactococci - a review." Syst. Appl. Microbiol. 10: 1- 19. 

Schleifer, K. H., and R. Kilpper-Balz. (1984). "Transfer of Streptococcus faecalis and Streptococcus faecium to the genus Enterococcus nom. rev. as Enterococcus faecalis comb. nov. and Enterococcus faecium comb. nov." Int. J. Syst. Bacteriol. 34: 31-34. 

Schleifer, K. H., Ehrmann, M., Krusch, U. and Neve, H. (1991). "Revival of the species Streptococcus thermophilus (ex Orla-Jensen, 1919) nom. rev." Syst. Appl. Microbiol. 14: 386- 388. 

Semedo, T., M. Almeida Santos, et al. (2003). "Comparative study using type strains and clinical and food isolates to examine hemolytic activity and occurrence of the cyl operon in enterococci." J Clin Microbiol 41(6): 2569-76. 

Shankar, N., A. S. Baghdayan, et al. (2002). "Modulation of virulence within a pathogenicity island in vancomycin-resistant Enterococcus faecalis." Nature 417(6890): 746-50. 

Shankar, N., P. Coburn, et al. (2004). "Enterococcal cytolysin: activities and association with other virulence traits in a pathogenicity island." Int J Med Microbiol 293(7-8): 609-18. 

Sharpe, M. E., L. R. Hill, et al. (1973). "Pathogenic lactobacilli." J Med Microbiol 6(3): 281-6. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 22 

Sidorchuk, II and V. M. Bondarenko (1984). "Selection of a biologically active mutant of Propionibacterium shermanii and the possibility of its use in complex therapy of enteral dysbacteriosis." J Hyg Epidemiol Microbiol Immunol 28(3): 331-8. 

Singh, K. V., S. R. Nallapareddy, et al. (2005). "Fsr-independent production of protease(s) may explain the lack of attenuation of an Enterococcus faecalis fsr mutant versus a gelE-sprE mutant in induction of endocarditis." Infect Immun 73(8): 4888-94. 

Slack, J. M. (1974). Genus I. Actinomyces Harz, 1877, 485. Bergey's Manual of Determinative Bacteriology. R. E. Buchanan, Gibbons, N.E. Baltimore, Williams & Wilkins Co.: 660-667. 

Tankovic, J., R. Leclercq, et al. (1993). "Antimicrobial susceptibility of Pediococcus spp. and genetic basis of macrolide resistance in Pediococcus acidilactici HM3020." Antimicrob Agents Chemother 37(4): 789-92. 

Tenorio, C., M. Zarazaga, et al. (2001). "Bifunctional enzyme 6'-N-aminoglycoside acetyltransferase-2"-O-aminoglycoside phosphotransferase in Lactobacillus and Pediococcus isolates of animal origin." J Clin Microbiol 39(2): 824-5. 

Torranzo, A., Magariños, B. and Romalde, J.E. (2005). "A review of the main bacterial fish diseases in mariculture systems." Aquaculture 246: 37-61. 

Torre, D., C. Sampietro, et al. (1990). "Necrotizing pneumonitis and empyema caused by Streptococcus cremoris from milk." Scand J Infect Dis 22(2): 221-2. 

Vagiakou-Voudris, E., D. Mylona-Petropoulou, et al. (2002). "Multiple liver abscesses associated with bacteremia due to Leuconostoc lactis." Scand J Infect Dis 34(10): 766-7. 

Wang, H. H., M. Manuzon, et al. (2006). "Food commensal microbes as a potentially important avenue in transmitting antibiotic resistance genes." FEMS Microbiol Lett 254(2): 226-31. 

Williams, A. M., J. L. Fryer, et al. (1990). "Lactococcus piscium sp. nov. a new Lactococcus species from salmonid fish." FEMS Microbiol Lett 56(1-2): 109-13. 

Wood, H. F., K. Jacobs, et al. (1955). "Streptococcus lactis isolated from a patient with subacute bacterial endocarditis." Am J Med 18(2): 345-7. 

Zeng, J., F. Teng, et al. (2005). "Gelatinase is important for translocation of Enterococcus faecalis across polarized human enterocyte-like T84 cells." Infect Immun 73(3): 1606-12. 

Zirakzadeh, A. and R. Patel (2006). "Vancomycin-resistant enterococci: colonization, infection, detection, and treatment." Mayo Clin Proc 81(4): 529-36. 

Appendix A - Assessment of gram-positive non-sporulating bacteria 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 23 

APPENDIX B. Scientific report on the assessment of Bacillus species 

For decades, strains belonging to several species of Bacillus have been deliberately introduced into the food chain either as plant protection products or as an animal feed supplement. The knowledge gained from this use suggests that, for some species at least, their safety could be assured by the “Qualified Presumption of Safety” (QPS) approach, according to EFSA (2005). 

1 – Identity of QPS unit. 

Since the first edition of the Bergey’s Manual of Systematic Bacteriology, the structure and content of the genus Bacillus have been substantially modified (Claus 1986; Ash, Priest et al. 1993; Priest 1993). In particular, several former Bacillus species have been excluded from the genus and reallocated to new genera. 

Most strains of Gram-positive, spore forming bacteria that have been or are used as animal feed supplements or plant protection products belong to species included in the new restricted definition of the genus Bacillus (Reva 2004; Hong, Duc le et al. 2005). The list of species included in this genus can be found in the second edition of the Bergey’s Manual of Systematic Bacteriology, Volume 3, or in http://www.bacterio.cict.fr/. 

The present document addresses species of the B. subtilis group that are, or were previously, classified as B. subtilis (B. amyloliquefaciens, B. atrophaeus, B. mojavensis, B. subtilis and B. vallismortis), and selected species within the B. cereus group (B. cereus, B. mycoïdes, B. pseudomycoides, B. thuringiensis and B. weihenstephanensis), since an extensive body of knowledge is available for these two groups either as the former species, or as the more recent, restricted species. Other species that that have been notified to EFSA, and for which a sufficient body of knowledge exists, are also considered (B. clausii, B. coagulans B. fusiformis, B. lentus B. licheniformis, B. megaterium, B. pumilus and Geobacillus stearothermophillus). 

The species listed above can be reliably identified using a 16S rRNA gene sequence. For species of the B. subtilis group the sequence of the gyraseA gene discriminates strains of the constituent species (Chun and Bae 2000). The species within the B. cereus group are difficult to distinguish reliably. However, B. cereus (including B. mycoides, B. pseudomycoides and B. weihenstephanensis) do not synthesise a parasporal crystal protein, and can thereby be distinguished from the crystalliferous B. thuringinesis. 

Appendix B - Assessment of Bacillus species 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 1 

2 - Is the body of knowledge sufficient? 

Several strains belonging to the species B. clausii, B. coagulans, B. licheniformis, B. megaterium, B. pumilus, B. subtilis and B. cereus (excluding B. anthracis) have been used as probiotics, animal feed supplements or in aquaculture (SCAN 2000; Hong, Duc le et al. 2005). Furthermore, several Bacillus species are involved in the preparation of traditional fermented dishes in Africa and Asia (Sarkar, Hasenack et al. 2002). Strains of B. thuringiensis, a species which belongs to the B. cereus group and is characterised by the production of insect toxic proteins which form a parasporal crystal in the bacterial cell, has a long history of use as plant protection products. Strains of B. cereus (Halverson and Handelsman 1991) and of B.subtilis (Krebs 1998; Kahn 2002; Cavaglieri, Orlando et al. 2005) have been used for treatment of seeds and roots to protect or promote the growth of plants. Several Bacillus species, as well as Geobacillus stearothermophilus, which does not grow below 45oC, are major sources of commercial enzymes (SCAN 2000). No foodborne cases or food safety problems have been linked to these usages of Bacillus strains so far, although it can be noted that B. thurigiensis strains, identical to those present in commercial plant protection products, have been isolated from faecal samples of greenhouse workers after exposure to such products (Jensen, Larsen et al. 2002). 

Annotated genome data are currently available for several strains within the species B. clausii, B. cereus, B. licheniformis, B. subtilis, B. thuringiensis, and Geobacillus kaustophilus thereby contributing significantly to the body of knowledge and decreasing the probability that unforeseen hazards could be associated with these bacilli (http://www.cbs.dtu.dk/services/GenomeAtlas/). 

3 - Are there safety concerns? 

3.1 Human safety concerns 

3.1.1 Anthrax 

The highly toxic bacterium Bacillus anthracis, the cause of anthrax in humans and animals, is a member of the Bacillus cereus group, and other members of the group have been seen to carry genes encoding anthrax-like toxins (Hoffmaster, Ravel et al. 2004). 

3.1.2 Food poisoning 

Several of the Bacillus species used as animal feed supplements, probiotics, plant protection products or seed coating agents are also known as agents of food poisoning (Kramer 1989; Granum 2000; EFSA 2005). 

1. cereus is a frequent agent of foodborne diseases. It is the cause of two distinct kinds of poisoning, (i) an emetic intoxication caused by the ingestion of the toxin cereulide, which is preformed in the food (Granum 2001), and (ii) a diarrhoeal infection caused by the production of either of the enterotoxins hemolysin BL (Hbl), non-hemolytic enterotoxin (Nhe) or 

Appendix B - Assessment of Bacillus species 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 2 

cytotoxin K (CytK) in the small intestine. Also B. thuringiensis has been implicated in cases of food poisoning (Jackson, Goodbrand et al. 1995) and most strains of B. thuringiensis produce the diarrhoeal B. cereus enterotoxins (Gaviria Rivera, Granum et al. 2000; Hansen and Hendriksen 2001). The diagnostic methods that are currently used to identify the cause of food poisoning often do not distinguish between B. cereus and B. thuringiensis. Therefore, some of the clinical cases attributed to B. cereus may actually have been caused by strains of B. thuringiensis. 

Strains of B. fusiformis, B. licheniformis, B. mojavensis, B. pumilus, and B. subtilis are known as rare causes of foodborne poisoning (Kramer 1989). A few strains among these species produce cytotoxins (Salkinoja-Salonen, Vuorio et al. 1999; From, Pukall et al. 2005). Strains of other species have been reported to produce B. cereus–like toxins, although the identification of the organisms in these studies is uncertain (B. amyloliquefaciens, B. circulans, B. lentimorbis, B. lentus, and B. megaterium). 

3.1.3 Other clinical implications 

1. cereus have been responsible for severe, although rare clinical infections in humans (Teyssou 1998): endophtalmitis (Callegan, Engelbert et al. 2002; Callegan, Cochran et al. 2006), necrotizing infections (Darbar, Harris et al. 2005), endocarditis (Cone, Dreisbach et al. 2005), bacteremia (Hernaiz, Picardo et al. 2003), osteomyelitis (Popykin 2002), septicemia (Matsumoto, Suenaga et al. 2000), pneumonia (Miller, Hair et al. 1997; Gray, George et al. 1999) and liver abscess (Latsios, Petrogiannopoulos et al. 2003). 

3.1.4 Antibiotic resistance 

Resistance to macrolides has been observed in several cases in species belonging to the genus Bacillus. The most commonly found mechanism is encoded by the erm(D) gene (Gryczan, Israeli-Reches et al. 1984; Hue and Bechhofer 1992; Kim, Choi et al. 1993), but the genetic location and potential transferability of this gene has not been determined. However, resistance genes present on extra-chromosomal elements of bacilli include plasmid-encoded erm(C), which has been identified in B. subtilis (Monod, Denoya et al. 1986), plasmid encoded tet(L), which has been found in B. stearothermophilus and B. subtilis (Hoshino, Ikeda et al. 1985; Sakaguchi, Amano et al. 1988), as well as mobilisable plasmid-encoded tetracycline resistance of the B. cereus group organisms (Battisti, Green et al. 1985). Considering that many examples of highly potent mechanisms of conjugative plasmid transfer have been described for the genus Bacillus (Andrup, Smidt et al. 1998; Thomas, Morgan et al. 2001; Poluektova, Fedorina et al. 2004), there is a high probability that such plasmids mobilise other plasmids that encode resistance. Additionally, in B. cereus the tet(M) tetracycline resistance determinant has been found on the conjugative transposon Tn916, which is known to have a very broad host-range (Agerso, Jensen et al. 2002), and also B. subtilis is known to harbour conjugative transposons that encode tetracycline resistance (Roberts, Pratten et al. 1999). 

Appendix B - Assessment of Bacillus species 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 3 

3.1.5 Livestock safety concerns 

1. cereus has been implicated in rare cases of bovine mastitis, bovine and ovine abortion (Parkinson, Merrall et al. 1999; Rowan, Caldow et al. 2003). B. licheniformis has been responsible for rare cases of bovine and ovine abortions (Agerholm, Willadsen et al. 1997; Rowan, Caldow et al. 2003). 

4 – Can the safety concern be excluded? 4.1 Human safety concerns 4.1.1 Anthrax 

1. anthracis, which is obviously not suited for QPS status, cannot be distinguished easily from the other members of the B. cereus group by 16S rRNA sequencing (Sacchi, Whitney et al. 2002). 

However, absence of genes encoding anthrax-like toxins in species belonging to the B. cereus group can be verified by PCR (Cheun, Makino et al. 2003; Ryu, Lee et al. 2003). Additionally, absence of expression of the toxins can be verified by IP injection in mice. 

4.1.2 Food poisoning 

Within the genus Bacillus, food poisoning is clearly linked to the production of emetic toxins or to the production of the diarrhoeal enterototoxins Nhe, Hbl and CytK (Granum 2001; EFSA 2005). 

Emetic toxins (cyclic peptides) are heat stable, resists pasteurisation and most industrial food thermal sterilisation processes, and are produced at temperature between 12 and 32oC (From, Pukall et al. 2005). They are formed by bacilli present in food products prior to ingestion. 

Production of the emetic toxin cereulide has generally been believed to be restricted to a group of very closely related strains within the B. cereus group (Ehling-Schulz, Svensson et al. 2005), but was recently found in also in B. weihenstephanensis (Thorsen, Hansen et al. 2006). The cereulide biosynthetic gene cluster is located on a large plasmid, which may be subject to lateral transfer (Hoton, Andrup et al. 2005; Ehling-Schulz, Fricker et al. 2006). However, Bacillus strains encoding the cereulide biosynthetic genes can be identified by PCR (Ehling-Schulz, Fricker et al. 2004). Additionally, production of cereulide can be rapidly detected in cultures grown on agar media by its ability to inhibit sperm mobility (Andersson, Mikkola et al. 1998; Haggblom, Apetroaie et al. 2002; From, Pukall et al. 2005). 

Emetic intoxication caused by Bacillus species other than B. cereus is very uncommon and the body of knowledge on toxigenic strains is not as large as for B. cereus. However, strains producing toxins with emetic activity can be identified by the same assays as used for B. cereus (From, Pukall et al. 2005). 

Appendix B - Assessment of Bacillus species 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 4 

Diarrhoeal enterotoxins are produced in the mammalian gut after ingestion of strains of certain Bacillus species. Formation of enterotoxins is largely restricted to the B. cereus group, but has occasionally been reported in other members of the genus Bacillus. 

Strains of B. cereus involved in diarrhoeal cases tend to produce high level of enterotoxins (Guinebretiere, Broussolle et al. 2002). The cytotoxic activity on epithelial cells of culture supernatants is therefore a good indication of their ability to cause foodborne poisoning (SCAN 2000). Strains that have the potential to cause diarrhoea can thus be identified by the toxicity of culture supernatants on Vero cells or other epithelial cell cultures (Sandvig and Olsnes 1982; SCAN 2000). The strains produce the B. cereus-like enterotoxins at 37° C, with the exception of B. weihenstephanensis, which produces the toxins only at <30° C. 

Even though the production of enterotoxins can thus be addressed and eventually excluded, the fact that strains within the B. cereus group are well known agents of food poisoning makes QPS inapplicable for this group of strains. Current knowledge shows that the vast majority of B. cereus strains are toxin producers, and thus cannot meet the qualifications required for all strains within the Bacillus genus. 

4.1.3 Other clinical implications 

When identified, the causes of the non-gastrointestinal B. cereus infections in human, described in section 3.1.3, were mostly wounds, trauma, intravenous drug usage (Callegan, Engelbert et al. 2002; Popykin 2002; Darbar, Harris et al. 2005). Some infections were nosocomial (Gray, George et al. 1999; Matsumoto, Suenaga et al. 2000) and some occurred in immuno-compromised patients (Motoi, Ishida et al. 1997; Cone, Dreisbach et al. 2005). Even though relation with foods or the food chain seems very unlikely, it is concluded that the potential risks associated with the use of B. cereus are too many for this strain to be included on the QPS list. 

The virulence factors involved in infection by Bacillus spp. outside of the gastrointestinal tract are not as well known as for food poisoning. Bacillus spp. other than B. cereus, in particular some strains of B. licheniformis, B. pumilus, and B. subtilis produce very active cyclic biosurfactants with large potential medicinal and industrial applications (Mulligan 2005; Rodrigues, Banat et al. 2006). These cyclic biosurfactants are also haemolytic (Dufour, Deleu et al. 2005) and could be cytotoxic. The role of cytotoxins and of cytotoxic biosurfactant molecules in infections occurring outside the gastrointestinal tract caused by Bacillus spp. other than Bacillus cereus should be investigated further. This might provide a mean to discriminate strains with infectious potential. 

4.1.4 Antibiotic resistance 

Safety concerns associated with acquired resistance genes can be excluded as described in the opinion of the Scientific Panel on Additives and Products or substances used in Animal Feed (http://www.efsa.europa.eu/en/science/feedap/feedap_opinions/993.html) 

Appendix B - Assessment of Bacillus species 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 5 

4.2 Livestock safety concerns 

The origin of mastitis and abortion caused by B. cereus has not been clearly established but does not seem to be linked to feed. In the case of abortion caused by B. licheniformis the origin of the infection is not identified. In one case, it was reported that placenta and the digestive tract content contained large number of B. licheniformis (Parvanta 2000). The body of knowledge available does not indicate that presence of these species in feed was a cause of infections outside of the gastrointestinal tract in animals. 

(Rowan, Caldow et al. 2003) observed that all strains of Bacillus spp. implicated in mastitis and abortion in animals were cytotoxic on the Hep-2 epithelial cell line. As also mentioned in the context of human clinical concerns, the role of cytotoxins and of cytotoxic biosurfactant molecules in non gastro-intestinal livestock infections caused by Bacillus spp. other than Bacillus cereus should be investigated further. 

4.3 Exceptions related to end-use 

Safety concerns may be excluded when the use of the strain does not imply its entry into the food chain. This is the case for strains used as seed coating agents where the probability that the bacterial cells are transferred to the edible part of the crop is very low. Bacillus spp, including toxigenic strains, are naturally abundant in the soil and restricted use of well- described Bacillus strains producing either enterotoxins or cyclic peptides as seed coating agents only should not represent a threat to the environment. 

It is likely that strains used as seed coating agents act in part by the production of compounds inhibiting plant pathogens, and such compounds may also have a toxic activity on epithelial cells or on sperm cells. Therefore, in the case of seed coating agents, excluding strains toxic to epithelial cells or to sperm cells is not necessary to ensure food safety and might be contradictory with the purpose of using such strains. 

5 - Species proposed for QPS 

In conclusion, it is proposed to include a number of Bacillus species notified to EFSA (B. clausii, B. coagulans B. fusiformis, B. lentus B. licheniformis, B. megaterium, B. pumilus and Geobacillus stearothermophillus), and species previously or presently classified as B. subtilis (B. amyloliquefaciens, B. atrophaeus, B. mojavensis, B. subtilis and B. vallismortis) on the list of QPS granted units due to the substantial body of knowledge available about these bacteria. Since all bacteria within the listed species potentially possess toxigenic traits, absence of toxigenic activity needs to be verified for qualification. 

Bacillus spp. belonging to the Bacillus cereus sensu lato group (B. cereus sensu stricto, B. mycoïdes, B. pseudomycoides, B. thuringiensis and B. weihenstephanensis) are not proposed for QPS, since it is known that the vast majority of strains within of this group are toxin producers and thus not meet the required qualifications. 

Appendix B - Assessment of Bacillus species 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 6 

REFERENCES 

Agerholm, J. S., C. M. Willadsen, et al. (1997). "Diagnostic studies of abortion in Danish dairy herds." Zentralbl Veterinarmed A 44(9-10): 551-8. 

Agerso, Y., L. B. Jensen, et al. (2002). "The identification of a tetracycline resistance gene tet(M), on a Tn916-like transposon, in the Bacillus cereus group." FEMS Microbiol Lett 214(2): 251-6. 

Andersson, M. A., R. Mikkola, et al. (1998). "A novel sensitive bioassay for detection of Bacillus cereus emetic toxin and related depsipeptide ionophores." Appl Environ Microbiol 64(4): 1338-43. 

Andrup, L., L. Smidt, et al. (1998). "Kinetics of conjugative transfer: a study of the plasmid pXO16 from Bacillus thuringiensis subsp. israelensis." Plasmid 40(1): 30-43. 

Ash, C., F. G. Priest, et al. (1993). "Molecular identification of rRNA group 3 bacilli (Ash, Farrow, Wallbanks and Collins) using a PCR probe test. Proposal for the creation of a new genus Paenibacillus." Antonie Van Leeuwenhoek 64(3-4): 253-60. 

Battisti, L., B. D. Green, et al. (1985). "Mating system for transfer of plasmids among Bacillus anthracis, Bacillus cereus, and Bacillus thuringiensis." J Bacteriol 162(2): 543-50. 

Callegan, M. C., D. C. Cochran, et al. (2006). "Virulence factor profiles and antimicrobial susceptibilities of ocular bacillus isolates." Curr Eye Res 31(9): 693-702. 

Callegan, M. C., M. Engelbert, et al. (2002). "Bacterial endophthalmitis: epidemiology, therapeutics, and bacterium-host interactions." Clin Microbiol Rev 15(1): 111-24. 

Cavaglieri, L., J. Orlando, et al. (2005). "Biocontrol of Bacillus subtilis against Fusarium verticillioides in vitro and at the maize root level." Res Microbiol 156(5-6): 748-54. 

Cheun, H. I., S. I. Makino, et al. (2003). "Rapid and effective detection of anthrax spores in soil by PCR." J Appl Microbiol 95(4): 728-33. 

Chun, J. and K. S. Bae (2000). "Phylogenetic analysis of Bacillus subtilis and related taxa based on partial gyrA gene sequences." Antonie Van Leeuwenhoek 78(2): 123-7. 

Claus, D., Berkeley, R.C.W. (1986). Genus Bacillus Cohn 1872, 174AL. Bergey's Manual of Systematic Bacteriology. P. H. A. Sneath, Mair, N.S., Sharpe, M.E., Holt, J.G. Baltimore, Williams & Wilkins: 1105-1139. 

Cone, L. A., L. Dreisbach, et al. (2005). "Fatal Bacillus cereus endocarditis masquerading as an anthrax-like infection in a patient with acute lymphoblastic leukemia: case report." J Heart Valve Dis 14(1): 37-9. 

Darbar, A., I. A. Harris, et al. (2005). "Necrotizing infection due to Bacillus cereus mimicking gas gangrene following penetrating trauma." J Orthop Trauma 19(5): 353-5. 

Dufour, S., M. Deleu, et al. (2005). "Hemolytic activity of new linear surfactin analogs in relation to their physico-chemical properties." Biochim Biophys Acta 1726(1): 87-95. 

Appendix B - Assessment of Bacillus species 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 7 

EFSA (2005). "Opinion of the Scientific comittee on a request from EFSA related to a generic approach to the safety assessment by EFSA of microorganisms used in food/feed and the production of food/feed additives." The EFSA Journal 226: 1-12. 

EFSA (2005). "Opinion of the scientific panel on biological hazards on Bacillus cereus and other Bacillus spp in foodstuff." The EFSA Journal 175: 1-48. 

Ehling-Schulz, M., M. Fricker, et al. (2006). "Cereulide synthetase gene cluster from emetic Bacillus cereus: structure and location on a mega virulence plasmid related to Bacillus anthracis toxin plasmid pXO1." BMC Microbiol 6: 20. 

Ehling-Schulz, M., M. Fricker, et al. (2004). "Identification of emetic toxin producing Bacillus cereus strains by a novel molecular assay." FEMS Microbiol Lett 232(2): 189-95. 

Ehling-Schulz, M., B. Svensson, et al. (2005). "Emetic toxin formation of Bacillus cereus is restricted to a single evolutionary lineage of closely related strains." Microbiology 151(Pt 1): 183-97. 

From, C., R. Pukall, et al. (2005). "Toxin-producing ability among Bacillus spp. outside the Bacillus cereus group." Appl Environ Microbiol 71(3): 1178-83. 

Gaviria Rivera, A. M., P. E. Granum, et al. (2000). "Common occurrence of enterotoxin genes and enterotoxicity in Bacillus thuringiensis." FEMS Microbiol Lett 190(1): 151-5. 

Granum, P. E. (2001). Bacillus cereus. Food Microbiology: Fundamentals and Frontiers. M. P. Doyle, et al. Washington, D.C., ASM Press: 373-381. 

Granum, P. E., Baird-Parker, T.C. (2000). Bacillus species. The microbiological safety and quality of food. B. M. Lund, Baird-Parker, T.C., Gould, G.W. Gaithersburg, Aspen Publishers: 1029-1056. 

Gray, J., R. H. George, et al. (1999). "An outbreak of Bacillus cereus respiratory tract infections on a neonatal unit due to contaminated ventilator circuits." J Hosp Infect 41(1): 19- 22. 

Gryczan, T., M. Israeli-Reches, et al. (1984). "DNA sequence and regulation of ermD, a macrolide-lincosamide-streptogramin B resistance element from Bacillus licheniformis." Mol Gen Genet 194(3): 349-56. 

Guinebretiere, M. H., V. Broussolle, et al. (2002). "Enterotoxigenic profiles of food-poisoning and food-borne Bacillus cereus strains." J Clin Microbiol 40(8): 3053-6. 

Haggblom, M. M., C. Apetroaie, et al. (2002). "Quantitative analysis of cereulide, the emetic toxin of Bacillus cereus, produced under various conditions." Appl Environ Microbiol 68(5): 2479-83. 

Halverson, L. J. and J. Handelsman (1991). "Enhancement of soybean nodulation by Bacillus cereus UW85 in the field and in a growth chamber." Appl Environ Microbiol 57(9): 2767-70. 

Hansen, B. M. and N. B. Hendriksen (2001). "Detection of enterotoxic Bacillus cereus and Bacillus thuringiensis strains by PCR analysis." Appl Environ Microbiol 67(1): 185-9. 

Appendix B - Assessment of Bacillus species 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 8 

Hernaiz, C., A. Picardo, et al. (2003). "Nosocomial bacteremia and catheter infection by Bacillus cereus in an immunocompetent patient." Clin Microbiol Infect 9(9): 973-5. 

Hoffmaster, A. R., J. Ravel, et al. (2004). "Identification of anthrax toxin genes in a Bacillus cereus associated with an illness resembling inhalation anthrax." Proc Natl Acad Sci U S A 101(22): 8449-54. 

Hong, H. A., H. Duc le, et al. (2005). "The use of bacterial spore formers as probiotics." FEMS Microbiol Rev 29(4): 813-35. 

Hoshino, T., T. Ikeda, et al. (1985). "Nucleotide sequence of the tetracycline resistance gene of pTHT15, a thermophilic Bacillus plasmid: comparison with staphylococcal TcR controls." Gene 37(1-3): 131-8. 

Hoton, F. M., L. Andrup, et al. (2005). "The cereulide genetic determinants of emetic Bacillus cereus are plasmid-borne." Microbiology 151(Pt 7): 2121-4. 

Hue, K. K. and D. H. Bechhofer (1992). "Regulation of the macrolide-lincosamide- streptogramin B resistance gene ermD." J Bacteriol 174(18): 5860-8. 

Jackson, S. G., R. B. Goodbrand, et al. (1995). "Bacillus cereus and Bacillus thuringiensis isolated in a gastroenteritis outbreak investigation." Lett Appl Microbiol 21(2): 103-5. 

Jensen, G. B., P. Larsen, et al. (2002). "Bacillus thuringiensis in fecal samples from greenhouse workers after exposure to B. thuringiensis-based pesticides." Appl Environ Microbiol 68(10): 4900-5. 

Kahn, M. R., Kahn, S.M. (2002). "Effects of root-dip treatment with certain phosphate solubilizing microorganisms on the fusarial wilt of tomato." Bioresource technology 85: 213- 215. 

Kim, H. S., E. C. Choi, et al. (1993). "A macrolide-lincosamide-streptogramin B resistance determinant from Bacillus anthracis 590: cloning and expression of ermJ." J Gen Microbiol 139(3): 601-7. 

Kramer, J. M., Gilbert, R.J. (1989). Bacillus cereus and other Bacillus species. Foodborne bacterial pathogens. New York, Marcel Dekker Inc. 

Krebs, B. N., Hoding, B., Kubart, S., Workie, M.A. Junge, H., Schmiedeknecht, G., grosch, R., Bochow, H., Hevesi, M. (1998). "Use of Bacillus subtilis as biocontrol agents". In: "Activity and characterization of Bacillus subtilis strains." Journal of Plant Disease Protection 105: 181-197. 

Latsios, G., C. Petrogiannopoulos, et al. (2003). "Liver abscess due to Bacillus cereus: a case report." Clin Microbiol Infect 9(12): 1234-7. 

Matsumoto, S., H. Suenaga, et al. (2000). "Management of suspected nosocomial infection: an audit of 19 hospitalized patients with septicemia caused by Bacillus species." Jpn J Infect Dis 53(5): 196-202. 

Appendix B - Assessment of Bacillus species 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 9 

Miller, J. M., J. G. Hair, et al. (1997). "Fulminating bacteremia and pneumonia due to Bacillus cereus." J Clin Microbiol 35(2): 504-7. 

Monod, M., C. Denoya, et al. (1986). "Sequence and properties of pIM13, a macrolide- lincosamide-streptogramin B resistance plasmid from Bacillus subtilis." J Bacteriol 167(1): 138-47. 

Motoi, N., T. Ishida, et al. (1997). "Necrotizing Bacillus cereus infection of the meninges without inflammatory reaction in a patient with acute myelogenous leukemia: a case report." Acta Neuropathol (Berl) 93(3): 301-5. 

Mulligan, C. N. (2005). "Environmental applications for biosurfactants." Environ Pollut 133(2): 183-98. 

Parkinson, T. J., M. Merrall, et al. (1999). "A case of bovine mastitis caused by Bacillus cereus." N Z Vet J 47(4): 151-2. 

Parvanta, M. F. (2000). "Abortion in a dairy herd associated with Bacillus licheniformis." Tierarztliche Umschau 55: 126-+. 

Poluektova, E. U., E. A. Fedorina, et al. (2004). "Plasmid transfer in bacilli by a self- transmissible plasmid p19 from a Bacillus subtilis soil strain." Plasmid 52(3): 212-7. 

Popykin, A., Blumberg, H.M., Hampton, R.W., Kripalani, S. (2002). "Osteomyelitis from Bacillus cereus infection following a gunshot injury." Journal of General Internal Medicine 17: 63. 

Priest, F. G. (1993). Systematics and ecology of Bacillus. Bacillus subtilis and Other Gram- Positive Bacteria. A. L. Sonenshein, Hoch, J.A., Losick, R. Washington D.C., American Society for Microbiology: 3-16. 

Reva, O. N., Dixelius, C., Meijer, J., Priest, F.G. (2004). "Taxonomic characterization and plant colonizing abilities of some bacteria related to Bacillus amyloliquefaciens and Bacillus subtilis." FEMS Microbiology ecology 48: 249-259. 

Roberts, A. P., J. Pratten, et al. (1999). "Transfer of a conjugative transposon, Tn5397 in a model oral biofilm." FEMS Microbiol Lett 177(1): 63-6. 

Rodrigues, L., I. M. Banat, et al. (2006). "Biosurfactants: potential applications in medicine." J Antimicrob Chemother 57(4): 609-18. 

Rowan, N. J., G. Caldow, et al. (2003). "Production of diarrheal enterotoxins and other potential virulence factors by veterinary isolates of bacillus species associated with nongastrointestinal infections." Appl Environ Microbiol 69(4): 2372-6. 

Ryu, C., K. Lee, et al. (2003). "Sensitive and rapid quantitative detection of anthrax spores isolated from soil samples by real-time PCR." Microbiol Immunol 47(10): 693-9. 

Sacchi, C. T., A. M. Whitney, et al. (2002). "Sequencing of 16S rRNA gene: a rapid tool for identification of Bacillus anthracis." Emerg Infect Dis 8(10): 1117-23. 

Appendix B - Assessment of Bacillus species 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 10 

Sakaguchi, R., H. Amano, et al. (1988). "Nucleotide sequence homology of the tetracycline- resistance determinant naturally maintained in Bacillus subtilis Marburg 168 chromosome and the tetracycline-resistance gene of B. subtilis plasmid pNS1981." Biochim Biophys Acta 950(3): 441-4. 

Salkinoja-Salonen, M. S., R. Vuorio, et al. (1999). "Toxigenic strains of Bacillus licheniformis related to food poisoning." Appl Environ Microbiol 65(10): 4637-45. 

Sandvig, K. and S. Olsnes (1982). "Entry of the toxic proteins abrin, modeccin, ricin, and diphtheria toxin into cells. I. Requirement for calcium." J Biol Chem 257(13): 7495-503. 

Sarkar, P. K., B. Hasenack, et al. (2002). "Diversity and functionality of Bacillus and related genera isolated from spontaneously fermented soybeans (Indian Kinema) and locust beans (African Soumbala)." Int J Food Microbiol 77(3): 175-86. 

SCAN (2000). "Opinion of the Scientific Committee on Animal Nutrition on the Safety of use of Bacillus species in animal nutrition." 

Teyssou, R., Hance, P., Buisson, Y. (1998). "Les infections humaines à Bacillus." Bulletin de la Société Française de Microbiologie 13: 137-144. 

Thomas, D. J., J. A. Morgan, et al. (2001). "Plasmid transfer between Bacillus thuringiensis subsp. israelensis strains in laboratory culture, river water, and dipteran larvae." Appl Environ Microbiol 67(1): 330-8. 

Thorsen, L., B. M. Hansen, et al. (2006). "Characterization of emetic Bacillus weihenstephanensis, a new cereulide-producing bacterium." Appl Environ Microbiol 72(7): 5118-21. 

Appendix B - Assessment of Bacillus species 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 11 

APPENDIX C: Scientific report on the assessment of Yeasts 

Today, the impact of yeasts on food and beverage production extends beyond the original and popular notions of bread, beer and wine fermentations by Saccharomyces cerevisiae (Querol, Belloch et al. 2003; Fleet 2006). Yeasts contribute to the fermentation of a broad range of other commodities, where various species may work in concert with bacteria and/or filamentous fungi. Many valuable food ingredients and processing aids are now derived by exploiting yeast properties, such as anti- fungal activity, enabling the yeasts to be used as novel agents in the biocontrol of food spoilage. The probiotic activity of some yeasts is another novel property that is of increasing interest especially in relation to animal feed. Finally, there are environmental aspects to be considered when yeasts are used as biocontrol agents. 

Although yeasts are part of the microbiota of many foods and beverages they are rarely (if ever) associated with outbreaks or cases of food-borne illness. 

1. Identity of the QPS unit 

In addition to S. cerevisiae, S. bayanus and S. pastorianus, it is now well established that various species of Candida, Hanseniaspora, Issatchenkia, Kluyveromyces, Metschnikowia, Pichia and Schizosaccharomyces can make a positive contribution in the manufacture of fermented foods, dairy products, meats, cereals, coffee and sauces. The most frequently encountered and important species in dairy products are Debaryomyces hansenii, Yarrowia lipolytica, Kluyveromyces marxianus, S. cerevisiae, Galactomyces geotrichum, Candida celanoides and various Pichia species. In the case of the fermentation of meat sausages and maturation of hams various species of Debaryomyces, Yarrowia lipolytica and various Candida species are involved. S. cerevisiae, S. exiguous, C. humicola, C. milleri, C. kruseii, C. orientalis, Torulaspora delbrueckii and various Pichia species are used in the fermentation of cereal products. The growth and activities of a diversity of Hanseniaspora, Candida, Pichia, Issatchenkia, Kluyveromyces and Saccharomyces species have been reported in the fermentation of coffee beans and cocoa beans. Zygosaccharomyces rouxii, C. versatilis and C. etchellsii are important osmotolerant species that play a key role in soy sauce fermentation (Boekhout 2003; Romano 2006). Moreover, baker’s and brewer’s yeasts (S. cerevisiae) have been available for many years as dietary supplements because of their high contents of B-group vitamins, proteins, peptides, amino acids and trace minerals. 

In summary, some species of the genera Candida, Debaryomyces, Hanseniaspora, Kluyveromyces, Pichia, Saccharomyces, Schizosaccharomyces, and Xanthophyllomyces can be included as the most relevant and commonly encountered yeasts 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 1 

1.1. Taxonomy. 

The identification, naming and placing of yeasts in their proper evolutionary framework are of importance to many areas of science including the food industry. At present, approximately 750 yeast species are recognised but only a few are frequently isolated. Several definitions have been used to describe the yeast domain. Thus, yeasts may be defined as being ascomycetous or basidiomycetous fungi that reproduce vegetatively by budding or fission, with or without pseudohyphae and hyphae, and forming sexual states that are not enclosed in fruiting bodies (Boekhout 2003). In the last version of the “The yeasts; a taxonomic study” (Kurtzman 1999) there is an extensive taxonomic revision. 

1.2. Methods for yeast identification. 

Yeasts are commonly identified either phenotypically or, more recently, from diagnostic sequences. Methods based on phenotype, include fermentation reactions on a selected set of sugars and growth responses on various carbon and nitrogen sources or on other diagnostic compounds (Krejer-van Rij 1984; Barnett 2000). However, these characteristics can vary according to growth conditions and sometimes the species are defined by a unique physiological characteristic that is controlled by a single gene. By contrast, techniques using molecular biology are seen as an alternative to traditional methods since they analyse the genome independently of the physiological state of the cell (for a review see Boekhout 2003; Fernández-Espinar 2006). The nucleotide sequences of the domains D1 and D2 located at the 5' end of gene 26S (Kurtzman and Robnett 1998) and PCR amplification of ribosomal DNA regions and restriction of the gene 5.8S rRNA gene and the adjacent intergenic regions ITS1 and ITS2 are the molecular methods commonly used for the identification of yeasts (Fernández-Espinar 2006). These techniques are more reproducible and faster that the conventional methods based on physiological and morphological characteristics. 

2. Is the body of knowledge sufficient? 

Indigenous, also referred to as traditional, fermented foods are those popular products that since early history have formed an integral part of the diet and that can be prepared in the home or by cottage industries using relatively simple techniques and equipment. Some of these products have undergone industrial development and are now manufactured on a large scale. Yeasts occur in a wide range of fermented foods, made from ingredients of plant or animal origin. When yeasts are abundant they have a significant impact on food quality parameters such as taste, texture, odour and nutritional value. Although several products are obtained by natural fermentation, the use of traditional starter cultures is widespread. 

The principal yeasts pathogenic for humans are Candida albicans and Cryptococcus neoformans which cause a range of mucocutaneous, cutaneous, respiratory, central nervous, systemic and organ infections (Hazen 2003; Richardson 2003). Usually, healthy, immunocompetent individuals are not at 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 2 

risk of such infections. Generally, individuals with weakened health and immune function are at greatest risk, and include cancer and AIDS patients, hospitalised patients and patients who are administered immunosuppressive drugs, broad-spectrum bacterial antibiotics and radio- and chemotherapies. The increased frequency of such individuals in the community, has led to an increase in the reporting of yeast infections due to an increasing number of yeast species other than C. albicans and Cryp. Neoformans (Hazen 1995; Georgiev 2003; Hobson 2003; Richardson 2003). These include species that are frequently found in food such as Candida krusei/orientalis, P. anomala, Kluy. Marxianus, S. cerevisiae and various Rhodotorula species (Richardson 2003; Enache-Angoulvant and Hennequin 2005; Fleet 2006). 

2.1. Candida 

The genus Candida is the largest in number of species of the yeast genera, and is present in almost every environment. Yeasts of this genus are abundantly distributed in nature on land and sea, associated with animals or plants and inanimate objects. 

2.1.1. Taxonomic unit defined 

The genus Candida comprises 163 species; including anamorphic ascomycetous yeasts that reproduce by multilateral budding and which are not assigned to morphological unique genera. As a result this is a highly heterogeneous genus comprising species whose perfect states are still unknown. This genus is distributed across the ascomycetous yeast domain, overlapping with other genera according to phylogenetic analysis using ribosomal genes (Kurtzman and Robnett 1998). 

This genus regroups a large number of species but the list can be refined to around 60 species that are present in food, the majority of them as spoilage organisms. A smaller number of species are used for food processing, as biocontrol agents (e.g. C. glabrata is used to control filamentous fungi in plants), or that are likely to be used by the industry and may cause opportunistic infections in humans. No literature pertaining to the use of Candida for animal feed can be found. 

The best targets for the identification of species of this genus are the sequences D1/ D2 (26S) or PCR and restriction of 5.8S-ITS 

The genome sequences are available for: 

- C. albicans (http://genolist.pasteur.fr/CandidaDB) - C. glabrata (http://cbi.labri.fr/Genolevures/elt/CAGL) - C. guilliermondii (http://www.broad.mit.edu/) - C. dubliniensis (http://www.genedb.org/genedb/cdubliniensis/) - C. lusitaniae (http://www.broad.mit.edu/) - C. parapsilosis (http://www.sanger.ac.uk/sequencing/Candida/parapsilosis/). - C. tropicalis (http://www.broad.mit.edu/) 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 3 

2.1.2. Is the body of knowledge sufficient? 

Species of this genus can be found in food-processing environments and have been recovered as contaminants in a large number of foods, such as fruits, fruit juices, soft drinks, alcoholic beverages, products with high sugar content, vegetables and grains, salted and acid preserved foods, dairy products, meat and meat-derived products. The list of species that are commonly used in the food industry include: C. zelanoydes, which contributes to the flavour and texture during the maturation of cheese and in the production of fermented milks (kefir and koumiss), C. milleri for flavour and rheology in sourdough breads, C. tropicalis, C. parapsilopsis and C. pelliculosa, which occur in the wet fermentation of coffee, C. etchellsii and C. versatilis, which contribute to the flavour of soy sauce, C. rugosa, which is involved in cocoa fermentations, C. utilis (=P. jadinii) and C. maltosa, which are used for biomass production from carbohydrate and hydrocarbon substrates respectively, C. oleophila and C. sake, which are commercialised for use as fungal biocontrol agents. 

2.1.3. Are there safety concerns? 

Humans. The principal human pathogenic yeasts are species of Candida, such as C. albicans, C. glabrata, C. guilliermondii, C. krusei, C. lusitaniae, C. parasilopsis, C. tropicalis, C. viswanathii (Richardson 2003). The principal and most common pathogen in the genus is C. albicans; other Candida species are however actually considered as emerging pathogens such as C. glabrata, which is being reported by some European medical centres as the main cause of candidaemia, C. parapsilosis, which is frequently isolated from skin lesions, and C. tropicalis, which is the second most frequent yeast, after C. albicans, causing deep-seated mycosis disease. Recently, new emerging pathogenic yeasts, such as C. dubliniensis, have been described, associated with hyperalimentation, broad-spectrum antibiotics, and immunosuppressive or antineoplastic therapies. Increasingly, systemic infections caused by C. guilliermondii are being reported. In most cases, invasive Candida infection is thought to be endogenous in origin, but transmission of organisms from person to person can also occur. 

Livestock. No information could be found regarding the use of Candida as animal feeds or probiotics. 

2.1.4. Can the safety concerns be excluded? 

The two main species of concern are C. albicans and C. tropicalis. Candida is involved in many cases of septicaemia (e.g. after surgery). 

2.1.5. List of units proposed for QPS status 

Although a number of Candida species associated with food appear not to cause infection, viz. C. etchellsii, C. maltosa, C. milleri, C. oleophila, C. pelliculosa, C. rugosa, C. sake, C. utilis (=P. 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 4 

jadinii), C. versatilis and C. zelanoydes, the fact that the principal human pathogenic yeasts are species of Candida, and that more and more Candida species are today considered as emergent pathogens, makes the genus Candida unsuitable for QPS status. 

2.2. Debaryomyces 

The genus Debaryomyces comprises 15 species. Many representatives can be isolated from natural habitats such as air, soil, pollen, tree exudates, plants, fruits, insects, and faeces and gut of vertebrates. 

2.2.1. Taxonomic unit defined 

Nine of these Debaryomyces species: D. carsonii, D. etchellsii, D. hansenii, D. maramus, D. melissophilus, D. polymorphus, D. pseudopolymorphus, D. robertsiae and D. vanrijiae, have been found in a variety of processed foods; such as fruit juices and soft drinks, wine, beer, sugary products, bakery products, dairy products and meat or processed meats. The presence of Debaryomyces species in foods usually has no detrimental effects and in some cases is beneficial to the food. The taxonomy is well defined with precise description available. The best method for species identification is PCR-RFLP of the IGS region of rDNA (Quiros, Martorell et al. 2006). 

The partial genome sequence of D. hansenii is available at (http://cbi.labri.fr/Genolevures/about/GL1_genome.php) 

2.2.2. Is the body of knowledge sufficient? 

Humans. Some Debaryomyces species are important in the ripening of fermented food products such as cheese and meat products. Where D. hansenii is used in the ripening of cheeses they metabolise lactic acid, raising the pH to allow the growth of proteolytic bacteria, and the yeast exhibits lipolytic activity that contributes to the development of cheese aromas. Proteolytic and lipolytic activities of D. hansenii have been described in the curing of ham and ripening of sausages and their presence in salami influences the red coloration and improves the quality of the product. Nevertheless, excessive growth of Debaryomyces species may cause undesirable sensory changes due to the formation of off- odours and off-flavours. These species have also been found as frequent contaminants of spoiled yoghurts, ice creams, fish, shellfish, etc. 

Livestock. No information could be found regarding the use of Debaryomyces as animal feeds or probiotics. 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 5 

2.2.3. Are there safety concerns? 

The main species of Debaryomyces used in food processing is D. hansenii, the anamorph form of which is Candida famata. C. famata has been repeatedly associated with catheter-related bloodstream infections, and occasionally with infections of the central nervous system. The reservoir of C. famata is not known but there is a possibility that nosocomial infections can occur via air contamination (Wagner, Sander et al. 2005). 

No studies on antifungal susceptibility of Debaryomyces are available. 

2.2.4. List of units proposed for QPS status 

It is proposed to grant D. hansenii QPS status. 

2.3. Hanseniaspora 

Hanseniaspora species are mainly found in the soil, on fruits and trees and in spoiled foods and beverages. Members of this genus are characterised by apiculate cells with vegetative reproduction by bipolar budding in basipetal succession. The six species in the genus that have valid descriptions are physiologically very similar; they ferment glucose, assimilate a few carbon compounds (arbutin, cellobiose, glucose, glucono-δ-lactone and salicin), and require inositol for growth. However, they show marked differences in the shape and number of ascospores, a criterion used for species identification. 

2.3.1. Taxonomic unit defined 

According to ribosomal sequences (26 S and 5.8S genes) the six species included in this genus are monophyletic and can be divided into two subgroups. This subdivision was supported by electrophoretic chromosome patterns. Hanseniaspora guilliermondii, H. uvarum and H. valbyensis have 8 to 9 chromosomes, while the second group comprises the species H. occidentalis, H. osmophila and H. vineae that have only 5 chromosomes. The anamorphic form of this genus is the well known Kloeckera. 

The best targets for identifying species of this genus are the sequences D1/ D2 (26S) or PCR and restriction of 5.8S-ITS 

2.3.2. Is the body of knowledge sufficient? 

Humans. The species are most frequently isolated from soil, fruits and plant exudates. The species occur on grapes and processed fruit. H. uvarum, the most relevant species of this genus, is important 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 6 

in the first phase of grape fermentation and is supposed to play a role in the production of certain flavours beneficial for the quality of wine and cider. Little is known regarding the other species. 

Livestock. No information could be found regarding the use of Hanseniaspora as animal feeds or probiotics. 

2.3.3. Are there safety concerns? 

No data concerning this yeast causing opportunistic infections have been found. 

2.3.4. List of units proposed for QPS status 

1. uvarum is proposed for QPS status 

2.4. Kluyveromyces 

The genus Kluyveromyces has a very wide distribution; its representatives have been recovered from substrates as diverse as seawater, soil, insects, plant material, fresh fruit, jams and other fruit preserves, dairy and bakery products, and have also been isolated from breweries, wineries and mammalian sources. Of the six species present in this genus, the most important are K. lactis and K. marxianus (anamorph C. kefyr) for their capacity to ferment lactose. 

2.4.1. Taxonomic unit defined 

The investigation of the phylogenetic relationships among the members of the genus Kluyveromyces has revealed the existence of a monophyletic group that is the new Kluyveromyces genus, based on multigene sequences analysis (Kurtzman 2003). This genus is constituted by the species K. aestuarii, K. dobzhanskii, K. lactis, K. marxianus, K. nonfermentans and K. wickerhamii. 

The best targets for identifying species of this genus are the sequences D1/ D2 (26S) or PCR and restriction of 5.8S-ITS. 

The partial genome sequences are available for the following Kluyveromyces species: - K. lactis (http://cbi.labri.fr/Genolevures/elt/KLLA) - K. marxianus (http://cbi.labri.fr/Genolevures/about/GL1_genome.php) 

2.4.2. Is the body of knowledge sufficient? 

Humans. This microorganism can be isolated from milk products and is used as a starter to set up the medium for cheese and kefir production. Kluyveromyces marxianus and K. lactis are associated with 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 7 

smear-ripened cheeses and contribute to the aromas that cheeses develop. These species are considered to be generally regarded as safe organisms and have been approved as a food additive (Coenen, Bertens et al. 2000). 

Livestock. Kluyveromyces is used in animal feeds in Europe as a probiotic and is apparently safe (reviewed in Anadon, Martinez-Larranaga et al. 2006). 

3.4.3. Are there safety concerns? 

Candida kefyr, the anamorph of K. marxianus, has occasionally been involved in opportunistic infections in immunocompromised persons. However, considering the history of apparent safe use and the rarity of infections in humans, there are no safety concerns. 

3.4.5. List of units proposed for QPS status 

It is proposed to grant K. lactis and K. marxianus QPS status. 

3.5. Pichia 

Yeasts of the genus Pichia are widely distributed; they can be found in natural habitats, such as soil, freshwater, tree exudates, insects, plants and fruits, and also as contaminants in a variety of foods and beverages, including juices and soft drinks, alcoholic beverages, high sugar containing products, vegetables, meat and fermented products. Moreover, some Pichia species have also been found exhibiting desired effects in food, e.g. contributing in the early stages of wine fermentation, several types of brines, and different types of cheeses; while others have been described as human pathogens (Bakir, Cerikcioglu et al. 2004; Otag, Kuyucu et al. 2005). 

3.5.1. Taxonomic unit defined 

The genus Pichia is one of the largest yeast genera in view of the number of species. Since the genus was described in 1904, the number of species included in this taxon has changed considerably. Pichia currently contains 91 species with 30 being related to food production and processing. However, the majority of them can be considered as food contaminants (spoilage organisms). Pichia appears to be extremely heterogeneous. The genus nearly doubled in size with the transfer of nitrate-positive Hansenula species to Pichia. For example the previously-named Hansenula polymorpha and H. angusta are now know as P. angusta, and H. anomala is now called P. anomala. 

The best targets for identifying species of this genus are the sequences D1/ D2 (26S) or PCR and restriction of 5.8S-ITS. 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 8 

The genome sequence of P. stipitis (http://genome.jgi-psf.org/cgi- bin/browserLoad/457559ad1ccd245f58d7b393) and partial genome sequences of P. angusta (http://cbi.labri.fr/Genolevures/about/GL1_genome.php) are available. 

3.5.2. Is the body of knowledge sufficient? 

Humans. The genus contains the species previously encompassed in the genus Hansenula, which is reported to be one of the safest microorganisms; it is used by the WHO for the development of vaccines and as a producer organism (e.g. phytases). The main species are P. anomala (previously Hansenula anomala) and P. angusta (previously Hansenula polymorpha). P. anomala is also used for the fermentation of bakery products, while P. roqueforti is used as a post-harvest biocontrol agent for wheat and barley, or for food application (olive fermentations). 

Pichia pastoris is frequently used as an expression system for the production of proteins. A number of properties makes Pichia suited for this task: Pichia has a high growth rate and is able to grow on a simple, inexpensive medium. Pichia can grow in either shake flasks or a fermentor, which makes it suitable for both small and large scale production. Pichia pastoris has a strong inducible promoter. This inducible promoter is related to the fact that Pichia pastoris is a methylotrophic yeast (Cereghino and Cregg 2000). 

Pichia jadinii (anamorph Candida utilis), commonly called Torula, in its inactive form (usually labelled as Torula yeast), is widely used as a flavouring in processed foods and pet foods. It is produced from wood sugars, as a by-product of paper production. It is pasteurized and spray-dried to produce a fine, light greyish-brown powder with a slightly yeasty odour and gentle, slightly meaty taste. 

Livestock. Some species of Pichia are used for feed (source of proteins), and as producer organisms (production of glucan for feed applications). 

3.5.3. Are there safety concerns? 

In the literature, P. anomala is described as a safe producer organism, since this yeast does not contain pyrogens or “viral inclusions” and is not a pathogen (Gellissen 2000). 

However, there is a single report of P. angusta and P. anomala being responsible for cases of fungaemia in a Brazilian paediatric intensive care unit (Pasqualotto, Sukiennik et al. 2005). The source of the infection was never found. Patients with P. anomala fungaemia seem to have risk factors in common with those who have candidaemia. A number of transient cases of candidaemia caused by Candida utilis (Pichia jadinii) have been reported. 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 9 

3.5.4. Can the safety concerns be excluded? 

Considering the history of safe use and the rarity of infections in humans, there are no safety concerns. 

3.5.5. List of units proposed for QPS status 

It is proposed that P. angusta , P. anomala, P. jadinii and P. pastoris have QPS status. 

3.6. Saccharomyces 

These species are strongly fermentative, and are commonly isolated from soil, fruits, foods and beverages. S. cerevisiae, S. pastorianus and S. bayanus are widely used for making bread and in the production of beer, wine, distilled beverages and fuel alcohol. S. cerevisiae occurs on fruit, in processed fruits, dairy products and plays a role in the fermentation of kefir, coffee, cocoa, and the production of traditional fermented products. S. cerevisiae and S. bayanus cause spoilage of soft drinks. 

3.6.1. Taxonomic unit defined 

According to (Kurtzman 1999), the genus Saccharomyces includes 14 species. However more recently, and based on multigene sequence analysis, (Kurtzman 2003) proposed a new Saccharomyces genus that includes only seven of the previous species (S. cerevisiae, S. paradoxus, S. mikatae, S. cariocanus, S. kudriavzevii, S. pastorianus and S. bayanus), the rest of the previous species are in a new genus, namely Kazachstania. The nucleotide sequence of the genome of each species has been determined but there are no annotations related to safety. Genome sequences are available for: 

- S. cerevisiae (http://mips.gsf.de/genre/proj/yeast/ and http://www.yeastgenome.org) - S. cerevisiae strain YJM789 isolated from the lung of fungal infections (http://www.ncbi.nlm.nih.gov/sites/entrez NC_009688) - S. bayanus (http://www.broad.mit.edu/annotation/fungi/comp_yeasts/ and http://cbi.labri.fr/Genolevures/about/GL1_genome.php - partial sequence) - S. mikatae (http://www.broad.mit.edu/annotation/fungi/comp_yeasts/) - S. paradoxus (http://www.broad.mit.edu/annotation/fungi/comp_yeasts/) 

The best targets for identifying species of this genus are the sequences D1/ D2 (26S) or PCR and restriction of 5.8S-ITS. 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 10 

3.6.2. Is the body of knowledge sufficient? 

Humans. The genus contains the most industrially exploited species known to man. It also contains the organism of choice as a model system for eukaryotic cell biology. As described above, this genus is used in the production of different foods, bread, beer, wine, distilled beverages and fuel alcohol, in processed fruits, dairy products and it plays a role in the fermentation of kefir, coffee, cocoa, and the production of traditional fermented products. Products are also commercialised as active dried yeast preparations. A subtype of S. cerevisiae (Saccharomyces boulardii) has been used by the food industry for many years as a probiotic, for horses as a treatment against acute enterocolitis, for human health against ulcerative colitis, in combination with standard antibiotics against Clostridium difficile disease, and as a treatment against persistent diarrhoea in children. S. cerevisiae is also used as biocontrol agent against Solanaceae diseases (Czerucka and Rampal 2002). 

Livestock. Saccharomyces is used in animal feeds in the European Union as a probiotic (reviewed in (Anadon, Martinez-Larranaga et al. 2006)). 

3.6.3. Are there safety concerns? 

Saccharomyces cerevisiae (also known as “baker’s yeast” or “brewers yeast”) is mostly considered to be an occasional digestive commensal. However, since the 1990’s, there have been a growing number of reports about its implication as an aetiological agent of invasive infection in “fragile” populations. A particular feature of such infections is their association with a probiotic preparation of S. cerevisiae (subtype S. boulardii) for treatment of various diarrhoeal disorders (see below). The nature of S. cerevisiae (subtype S. boulardii) and its clinical applications are reviewed by (Buts and Bernasconi 2005). 

In one review, 92 cases of Saccharomyces invasive infection were presented (Enache-Angoulvant and Hennequin 2005). Predisposing factors were similar to those of invasive candidosis, with intravascular and antibiotic therapy being the most frequent. Blood was the most frequent site of isolation (78% or 72 patients). S. cerevisiae (subtype S. boulardii) accounted for 51.3% (47 cases) of fungaemias and was exclusively isolated from blood. Special caution should be taken regarding the use of S. cerevisiae (subtype S. boulardii) preparations (Fleet and Roostita, 2006). There are number of recent reports and reviews regarding the safety of S. cerevisiae (subtype S. boulardii) preparations involved in: 

• A case of Saccharomyces cerevisiae acquired fungaemia (Cassone, Serra et al. 2003; Graf and Gavazzi 2007). The authors concluded that probiotics should be used cautiously in certain high-risk populations. 
• A review of the current literature reinforces the view that fungaemia and sepsis are rare complications of the administration of S. cerevisiae (subtype S. boulardii) in immunocompromised patients but confirms that the most important risk factor for S. 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 11 

cerevisiae fungaemia is the use of probiotics (Herbrecht and Nivoix 2005; Munoz, Bouza et al. 2005). This raises the question of the risk-benefit ratio of these agents in critically ill or immunocompromised patients who are likely to develop an infection after exposure to high amounts of a microorganism with a low virulence. 

The body of knowledge is considered as sufficient (long history of safe use) with only 92 cases of pathogenic cases involving S. cerevisiae reported in total (15 cases diagnosed before 1990); all patients had at least one condition facilitating the opportunistic development of S. cerevisiae. S. bayanus and S. pastorianus are used in wine and beer production. There are no foodborne infection issues for these species. 

3.6.4. Can the safety concerns be excluded? 

It is possible to propose some species of the genus for QPS status with the following qualification: S. cerevisiae, subtype S. boulardii should certainly be contraindicated for patients of fragile health, as well as for patients with a central venous catheter in place. It is recommended that a specific protocol concerning the use of probiotics needs to be formulated. 

3.6.5. List of units proposed for QPS status 

1. bayanus, S. cerevisiae and S. pastorianus (syn of S. carlsbergensis) are proposed for QPS status with the above qualification. 

3.7. Schizosaccharomyces 

Three species are included in this genus, Sch. japonicus, Sch. octosporus and Sch. pombe. The species are isolated from fruits and fruit juices, wines, tequila fermentation and high sugar concentration. The species are strong fermenters of sugars and have been used for the production of ethanol. 

The species Sch. pombe is used as a phytase producer for animal feed; minimum safety precautions should be taken for the handling and storage. No infection issues have been reported. 

The best targets for identifying species of this genus are the sequences D1/ D2 (26S) or PCR and restriction of 5.8S-ITS. 

The genome sequence for Sch. pombe is available (http://www.genedb.org/genedb/pombe/) 

3.7.1. List of units proposed for QPS status 

Sch. pombe is proposed for QPS status 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 12 

3.8. Xanthophyllomyces. 

Phaffia rhodozyma was isolated by Herman Phaff in the 1960s, during his pioneering studies of yeast ecology. The species ferments D-glucose and occurs in slime fluxes of deciduous trees. Initially, the yeast was isolated from limited geographical regions, but isolates were subsequently obtained from Russia, Chile, Finland, and the United States. The biological diversity of the yeast is more extensive than originally envisioned by Phaff and his collaborators, and at least two species appear to exist, including the anamorph Phaffia rhodozyma and the teleomorph Xanthophyllomyces dendrorhous. The yeast has attracted considerable biotechnological interest because of its ability to synthesize the economically important carotenoid astaxanthin (3,3’-dihydroxy-β, β-carotene-4,4’-dione) as its major pigment. This property has stimulated research on the biology of the yeast as well as development of the yeast as an industrial microorganism for astaxanthin production by fermentation. The pigment is an important dietary source for aquaculture and poultry industries, including salmonids, lobsters and the egg yolks of chickens and quail, in order to impart characteristic and desirable colours. 

The best targets for identifying species of this genus are the sequences D1/ D2 (26S). 

There are no literature reports to suggest that this species may be hazardous to human health. 

3.8.1. List of units proposed for QPS status 

Due to a long history of apparent safe use Xanthophyllomyces dendrorhous is proposed for QPS status 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 13 

REFERENCES 

Anadon, A., M. R. Martinez-Larranaga, et al. (2006). "Probiotics for animal nutrition in the European Union. Regulation and safety assessment." Regul Toxicol Pharmacol 45(1): 91-5. 

Bakir, M., N. Cerikcioglu, et al. (2004). "Pichia anomala fungaemia in immunocompromised children." Mycoses 47(5-6): 231-5. 

Barnett J.A., P. R. W., Yarrow D. (2000). Yeasts: Characteristics and identification, Cambridge University Press. 

Boekhout, T. a. R. V. (2003). Yeast in food. Beneficial and detrimental aspects. Cambridge, CRC Press. 

Buts, J. P. and P. Bernasconi (2005). "Saccharomyces boulardii: basic science and clinical applications in gastroenterology." Gastroenterol Clin North Am 34(3): 515-32, x. 

Cassone, M., P. Serra, et al. (2003). "Outbreak of Saccharomyces cerevisiae subtype boulardii fungemia in patients neighboring those treated with a probiotic preparation of the organism." J Clin Microbiol 41(11): 5340-3. 

Cereghino, J. L. and J. M. Cregg (2000). "Heterologous protein expression in the methylotrophic yeast Pichia pastoris." FEMS Microbiol Rev 24(1): 45-66. 

Coenen, T. M., A. M. Bertens, et al. (2000). "Safety evaluation of a lactase enzyme preparation derived from Kluyveromyces lactis." Food Chem Toxicol 38(8): 671-7. 

Czerucka, D. and P. Rampal (2002). "Experimental effects of Saccharomyces boulardii on diarrheal pathogens." Microbes Infect 4(7): 733-9. 

Enache-Angoulvant, A. and C. Hennequin (2005). "Invasive Saccharomyces infection: a comprehensive review." Clin Infect Dis 41(11): 1559-68. 

Fernández-Espinar, M. T., Martorell, P., de Llanos, R., Querol, A. (2006). Molecular methods to identify and characterise yeast in foods and beverages. The yeast handbook. A. Querol, Fleet, G.H. Berlin, Springer. 

Fleet, G. H. (2006). The commercial and community significance of yeasts in food and beverage production. The Yeast Handbook. A. Querol, Fleet, G.H. Berlin, Springer. 

Fleet, G. H., Roostita, R. (2006). The public health and probiotic significance of yeasts in foods and beverages. The Yeast Handbook. A. Querol, Fleet, G.H. Berlin, Springer. 

Gellissen, G. (2000). "Hansenula polymorpha - the ultimate protein machine." Helix 2: 6-8. 

Georgiev, V. S. (2003). Opportunistic Infections: Treatment and Prophylaxis. Totowa, New Jersey, Humana Press. 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 14 

Graf, C. and G. Gavazzi (2007). "Saccharomyces cerevisiae fungemia in an immunocompromised patient not treated with Saccharomyces boulardii preparation." J Infect 54(3): 310-1. 

Hazen, K. C. (1995). "New and emerging yeast pathogens." Clin Microbiol Rev 8(4): 462-78. 

Hazen, K. C., Howell, S.A. (2003). Candida, Cryptococcus and other yeasts of medical importance. Manual of Clinical Microbiology. P. R. Murray. Washington DC, ASM Press: 1693-1711. 

Herbrecht, R. and Y. Nivoix (2005). "Saccharomyces cerevisiae fungemia: an adverse effect of Saccharomyces boulardii probiotic administration." Clin Infect Dis 40(11): 1635-7. 

Hobson, R. P. (2003). "The global epidemiology of invasive Candida infections--is the tide turning?" J Hosp Infect 55(3): 159-68; quiz 233. 

Krejer-van Rij, N. J. W. (1984). The yeasts: a taxonomic study. Amsterdam, Elsevier Science Publishers B. V. 

Kurtzman, C. P. (2003). "Phylogenetic circumscription of Saccharomyces, Kluyveromyces and other members of the Saccharomycetaceae, and the proposal of the new genera Lachancea, Nakaseomyces, Naumovia, Vanderwaltozyma and Zygotorulaspora." FEMS Yeast Res 4(3): 233-45. 

Kurtzman, C. P., Fell, J.W. (1999). The yeasts: a taxonomic study. Amsterdam, Elsevier. 

Kurtzman, C. P. and C. J. Robnett (1998). "Identification and phylogeny of ascomycetous yeasts from analysis of nuclear large subunit (26S) ribosomal DNA partial sequences." Antonie Van Leeuwenhoek 73(4): 331-71. 

Munoz, P., E. Bouza, et al. (2005). "Saccharomyces cerevisiae fungemia: an emerging infectious disease." Clin Infect Dis 40(11): 1625-34. 

Otag, F., N. Kuyucu, et al. (2005). "An outbreak of Pichia ohmeri infection in the paediatric intensive care unit: case reports and review of the literature." Mycoses 48(4): 265-9. 

Pasqualotto, A. C., T. C. Sukiennik, et al. (2005). "An outbreak of Pichia anomala fungemia in a Brazilian pediatric intensive care unit." Infect Control Hosp Epidemiol 26(6): 553-8. 

Querol, A., C. Belloch, et al. (2003). "Molecular evolution in yeast of biotechnological interest." Int Microbiol 6(3): 201-5. 

Quiros, M., P. Martorell, et al. (2006). "PCR-RFLP analysis of the IGS region of rDNA: a useful tool for the practical discrimination between species of the genus Debaryomyces." Antonie Van Leeuwenhoek 90(3): 211-9. 

Richardson MD, W. D. (2003). Fungal Infection: Diagnosis and Management. Oxford, Blackwell Publishing. 

Richardson, M. D. W., D.W. (2003). Fungal Infection: Diagnosis and Management. Oxford, Blackwell Publishing. 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 15 

Romano, P., Capece, A., Jespersen, L. (2006). Taxonomic and ecological diversity of food and beverages yeasts. The yeast handbook. A. Querol, Fleet, G.H. Berlin, Springer. 

Wagner, D., A. Sander, et al. (2005). "Breakthrough invasive infection due to Debaryomyces hansenii (teleomorph Candida famata) and Scopulariopsis brevicaulis in a stem cell transplant patient receiving liposomal amphotericin B and caspofungin for suspected aspergillosis." Infection 33(5-6): 397-400. 

Appendix C - Assessment of Yeasts 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 16 

APPENDIX D: Scientific report on the assessment of filamentous fungi 

Filamentous fungi are flexible microorganisms that can show different properties depending on the external factors (substrate, growth conditions, biotic/abiotic conditions). A consequence is the ability of a microorganism to produce different types and quantity of secondary metabolites depending on the growth conditions. Mycotoxins are well-known secondary metabolites, and Penicillium roqueforti, for example, is safely used for cheese production but could start to produce a lot of mycotoxins if the substrate is changed, e.g. to rye bread. Regarding the QPS status, the ability of fungal species to produce toxic metabolites represents the greatest difficulties. Based on the assumption that each of the estimated 1.5 million fungal species (Hawksworth 1991) can produce at least two unique secondary metabolites, there may be as many as 3 millions unique fungal metabolites. Approximately 10% of the secondary metabolites listed up till now have been classified as mycotoxins. Thus, there are potentially up to 300,000 unique mycotoxins (CAST 2003). The number of fungal metabolites and mycotoxins still undiscovered is therefore quite large and the diversity of toxic mechanisms will be equally as great. There is, unfortunately, no standardised method to consider fungal metabolites and their toxicity such as effect-based bioassay methods. The regulation of metabolites and their possible interactions are therefore poorly understood. Just as for other types of microorganisms, the toxic effect of a fungus used for food production will only be detected in case of acute toxicity, but not if it shows long term (chronic) toxicity (e.g. carcinogenic properties). In addition, the number of validated analytical methods for mycotoxins and other fungal metabolites is low and even for those available, analytical quality assurance procedures are often lacking (van Egmond 2004). 

In contrast to bacteria, the spread of antimicrobial resistance through filamentous fungi is not a concern. Attention should, however, be paid to fungi capable of producing antibiotics that are not initially present in food and that therefore might contribute to the emergence of populations of resistant bacteria (e.g. penicillin production by some fungal species). 

A market-research study (Sunesen 2003) illustrates the difficulty of getting precise information on the identity of organisms used in food, and therefore to evaluate their safety. 

Mycological methods to be used for identification of moulds: 

Filamentous fungi are traditionally identified to genus level by phenotypic characters, such as morphological and cultural characteristics. Unfortunately, there is not one universal mycological textbook or reference compendium which is used for identification of moulds, which makes identification to genus level a highly subjective task. This is further complicated by the necessity to identify fungal strains to the species level as each species within a genus may have very different functional characters, e.g. mycotoxin profiles and physiological properties. Again, traditional methods like morphological and cultural characteristics are widely used but also profiles of secondary metabolites have been used within some genera. Phenotypic characteristics do vary according to growth conditions which makes it difficult to construct robust identification keys. No 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 1 

identification key covers all species, so it is recommended seeking advice for identification procedures by contacting specialists in food, feed and industrial mycology – e.g. via the International Commission on Food Mycology (ICFM) (http://www.foodmycology.org), which can direct inquires to recommended specialists. 

For filamentous fungi the use of molecular biology based methods is less developed than for bacteria and yeasts. On the other hand, in combination with phenotypic studies, numerous phylogenetic studies using gene sequences have changed the systematics within mycology and will play an increasing role in the future by changing our understanding of species delimitations and relationships. As a spin-out from molecular biology, some sequenced-based identification schemes have been developed (e.g. for Trichoderma – see below) along with various PCR detection systems. However, the latter systems often are intended for a limited number of species, at times only a minor part of a genus. The molecular methods developed so far are not based on the same gene(s) for different genera and need further improvement (Paterson 2006). Many recent phylogenetic studies and molecular detection systems are based on a Multi-Locus Sequence Typing (MLST) concept, where sequences from several genes are used simultaneously. Typical targets chosen for MLST typing are “housekeeping” genes, without which the host organism will be unable to function. Again, there will be differences among fungal genera regarding the loci used in MLST studies and advice should be obtained by consulting specialists – e.g. via the International Commission on Food Mycology. 

1. ASPERGILLUS 

This genus is among the best known filamentous fungi, as Aspergillus species are widely used for production of chemicals (e.g. citric acid), enzymes and for biotransformations. On the other hand, Aspergillus species are also known to be among the most toxic spoilers of food and feed, some species are even pathogenic to man and food producing animals. No recent monograph on Aspergillus and its teleomorphic states exists, which makes it complicated to give a clear picture of the current status within this genus where the systematics are currently changing rapidly. Specific reviews will be cited for the relevant species in the paragraphs that follow. For a general introduction to Aspergillus it may be useful to consult some general texts on food mycology (Pitt 1979; Samson 2004), as well as a recent update on concepts for species differentiation in Aspergillus (Samson 2006). 

1.1. Aspergillus section Nigri (the black Aspergilli) 

The taxonomy of the section Nigri (the black Aspergilli) is not fully resolved as the number of accepted species depends on the methodology used. So far there has not been complete agreement between morphological, chemical and molecular data, but some generally acceptance has been proposed (Schuster, Dunn-Coleman et al. 2002; Abarca, Accensi et al. 2004; Samson 2004); however species identification remains problematic. The section Nigri includes 16 species A. niger, A. foetidus, A. tubingensis, A. aculeatus, A. brasiliensis, A. carbonarius, A. costaricaensis, A. ellipticus, A. heteromorphus, A. homomorphus, A. ibericus, A. japonicus, A. lacticoffeatus, A. 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 2 

piperis, A. sclerotiniger, and A. vadensis; however only the first four species listed will be evaluated for a possible QPS status as they have been used for food or feed purposes, including enzyme production. 

1.2. Aspergillus niger 

In general Aspergillus niger sensu lato has a long history of apparent safe use in biotechnology, e.g. for the production of chymosin and other enzymes or citric acid (Schuster, Dunn-Coleman et al. 2002; van Dijck, Selten et al. 2003). A. niger is not known to be used as food or feed in Europe, even though this species has been evaluated for use as a source of single-cell protein (Christias, Couvaraki et al. 1975; Hang 1976; Singh, Abidi et al. 1991; Oboh 2002). Were a strain of A. niger to be allowed in Europe, it would fall under the Novel Food Regulation (258/97/EC) and would thus require a risk assessment under that legislation. 

The full nucleotide sequences of the genomes of three strains of Aspergillus niger sensu stricto have been determined and information are available at these web sites: http://genome.jgi-psf.org/Aspni1/Aspni1.home.html http://www.aspergillus.org.uk/indexhome.htm?secure/sequence_info/index.php~main 

It is well documented that some strains of this species produce the mycotoxin ochratoxin A (Abarca, Bragulat et al. 1994; Samson 2004; Serra, Cabanes et al. 2006). Other metabolites with poorly documented biological activity from A. niger are: pyranonigrin, kotanins and naphtho-γ- pyrones (Samson 2004). A. niger is the third most common species associated with invasive pulmonary aspergillosis and it is also often a causative agent of aspergilloma (Kwon-Chung 1992). It is also a recognised opportunistic pathogen for animals and there have been reports of natural aspergillosis in various species of mammals and birds (Smith 1989). 

Despite the long history of apparent safe use in biotechnology, where strain improvement combined with cleaning and purification steps have been added to processes to eliminate metabolites other than the product of interest (van Dijck, Selten et al. 2003; Blumenthal 2004), industrial strains of A. niger have been proven to produce ochratoxin A (Schuster, Dunn-Coleman et al. 2002) which makes A. niger ineligible for a QPS status. 

1.3. Aspergillus foetidus, Aspergillus tubingensis and Aspergillu aculeatus are each used for enzyme production. Even though A. aculeatus is from Section Nigri it can be distinguished morphologically from A. niger and the other species in the Nigri section (Pitt 1997). Nevertheless, due to the confused taxonomy of the Section Nigri in the past, many reports on enzyme production by A. niger should probably be attributed to isolates of A. foetidus, A. tubingensis or A. aculeatus. These species are known to produce many metabolites with poorly described biological activity. For A. foetidus, these include pyranonigrin, naphtho-γ-pyrones, asperazine, and anatafumicin (Samson 2004). A. tubingensis has been reported to produce pyranonigrin, naphtho-γ-pyrones, and asperazine, (Samson 2004), and A. foetidus to produce ochratoxin A (Teren, Varga et al. 1996; 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 3 

Bragulat, Abarca et al. 2001; Abarca, Accensi et al. 2004). One metabolite from A. aculeatus, secalonic acid, is known to be a mycotoxin (Samson 2004). 

Despite the long history of apparent safe use in biotechnology, the body of knowledge concerning the toxicological aspects of the metabolites is insufficient, which makes Aspergillus foetidus, A. tubingensis and A. aculeatus ineligible for a QPS status. 

1.4. Other Aspergilli 

1.4.1. Aspergillus candidus 

1. candidus can be found in meat products (sausages) as a starter culture with a long history of traditional use with regard to the house mycobiota (Sunesen 2003). This species is not produced commercially as starter culture (for application by spraying or dipping), hence it is not declared. A. candidus can also be found as a contaminant (food spoiler) in cereals and many other food products (Pitt 1997). 

Despite the frequent occurrence of this species the body of knowledge is considered as insufficient; it produces known metabolites, some of them showing cytotoxic activity: AcT1 (Chattopadhyay, Nandi et al. 1987), xanthoascin (Ito, Ohtsubo et al. 1978), terphenyllin (Marchelli and Vining 1975; Stead, Affleck et al. 1999). However, there remains metabolites that are not yet identified and classified (Samson 2004; Andersen and Thrane 2006). The toxicology of the metabolites of A. candidus is unknown, so the safety concerns cannot be excluded. Even if it is possible to get rid of most of the fungal biomass by washing the surface of the product, there is the possibility that fungal metabolites will remain on the product. Moreover, possible interactions between these metabolites have yet to be investigated. Some rare case of infections linked to A. candidus can be found in the literature (Kwon-Chung 1992; Ribeiro, Santana et al. 2005). 

In conclusion, considering that Aspergillus candidus is known to produce secondary metabolites with poorly understood toxicity for which there is no data on possible interactions, and that A. candidus is mainly used for food production as a house starter culture and therefore mixed with other fungi, makes A. candidus ineligible for a QPS status. 

1.4.2. Aspergillus oryzae 

In Asia a long tradition of using fungal cultures to produce fermented food such as sake (rice wine), shoyu (soy sauce) and miso (soybean paste) exists. These products are fermented by “koji-moulds”, which consist principally of Aspergillus oryzae, but may also contain A. “awamori” (=A. niger) , A. sojae and A. tamarii. The consumption of these fermented foods in Japan has been considered as safe (Tanaka 2006). Recent genomic approaches have demonstrated that A. oryzae and A. tamarii are taxonomically closely related to A. flavus, while A. sojae and A. awamori are genetically related to A. parasiticus and A. niger, respectively (Machida, Asai et al. 2005). 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 4 

Aspergillus oryzae has a long history of apparent safe use, both in food outside Europe (it is one of the main species used in Asia for the production of soy sauce, which is exported worldwide), and for enzyme / protein production (cell factory), however, this is as GM organisms (Archer 2000). A. oryzae is accepted as a domesticated form of A. flavus (Pitt 1997), which is an aflatoxin producer. The phenotypic distinction between A. oryzae and A. flavus is difficult as only fine details in conidial ornamentation and colony characteristics (i.e. colour of conidial mass and colour of colony reverse on Aspergillus Flavus Parasiticus Agar) separate the two (Samson 2004). However, by several molecular methods it has not been possible to separate the two into distinct species (Cary and Ehrlich 2006; Chang, Ehrlich et al. 2006). A. oryzae has the gene cluster for aflatoxin but has a minute change in the sequence for a regulatory gene, aflR, which is believed to the reason for the absence of aflatoxin production by A. oryzae (Lee, Liou et al. 2006). A recent review of the occurrence of aflatoxins and their production by various koji-moulds (Tanaka 2006) demonstrated that 212 strains used for fermentation of different foods were negative for aflatoxin production. Aflatoxins were not detected in any of the 289 food samples analysed (rice, soy sauce, soybean paste). 

Strains of A. ozyzae do, however, produce the mycotoxins cyclopiazonic acid, which is a neurotoxic and immunosuppressive compound, and β-nitropropionic acid and kojic acid (Samson 2004). Four of 36 A. oryzae-strains used commercially were found to be producers of cyclopiazonic acid (Goto 1987), whereas kojic acid was found to be produced by 85 of 149 koji-mould strains used commercially (Shinshi 1984). The strains producing toxin were removed from commercial use and Tanaka et al. (Tanaka 2006) concluded that the risk for mycotoxin contamination of typical Japanese fermented food can be classified as very low. A. oryzae is also used as feed for dairy cows and beef cattle in growth finishing stages; however, potential production of cyclopiazonic acid and β-nitropropionic acid were not taken into consideration (EFSA 2006). 

Despite the long history of apparent safe use in food and biotechnology, where cleaning and purification steps have been added in the process to get rid of all metabolites but the product of interest (Blumenthal 2004), the body of knowledge concerning the formation of well-known mycotoxins, cyclopiazonic acid and β-nitropropionic acid, under production conditions as well as any long-term toxicological aspects of these toxins is insufficient. In addition, no universally accepted method for an unambiguous identification of A. oryzae exists, which make Aspergillus oryzae not suitable for a QPS status. 

2. PENICILLIUM 

Among the most frequently encountered fungi in food and feed systems are species of the genus Penicillium, which are very well-known as spoilers and mycotoxin producers but also as starter cultures for products like e.g. white- and blue-mould cheeses and mould-ripened meat products. The modern systematics of the genus Penicillium was initiated by a monograph more than 25 years ago (Pitt 1979) and has developed dramatically since then. Today the genus is divided into four subgenera (Pitt 1997; Samson 2004) and may contain more than 500 species. Many species, however, are soil fungi and has never been related to food and feed systems, except as occasional spoilers. All Penicillium species are good producers of mycotoxins and other biological active 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 5 

metabolites, however the available literature is overwhelming and difficult to interpret as the identification of Penicillium cultures is not trivial and has resulted in numerous misidentifications (Frisvad, Nielsen et al. 2006). Partly as a consequence of this, starter cultures are often vaguely labelled as “Penicillium spores” (Sunesen 2003). Among the Penicillium strains used routinely in the food industry toxigenic strains are frequent. In a study of 249 Penicillium strains originally isolated from food products and used as starter cultures only 13 isolates were found to meet the demands on technological suitability and toxicological safety, which includes the testing of the strains with regard to the production of antibiotic, cytotoxic and mutagenic metabolites (Gareis 1999). 

Based on literature reviews, only species within the subgenus Penicillium have been used as starter cultures for food and feed. Recently this subgenus has been the subject of a monograph (Frisvad 2004; Samson 2004; Smedsgaard 2004) including an extensive review on the related secondary metabolites (Frisvad 2004). An interactive identification key based on phenotypic characters and β-tubulin gene sequences is available at: http://www.cbs.knaw.nl/penicillium/DefaultPage.aspx. 

2.1. Penicillium camemberti 

Penicillium camemberti has a long history of use in cheese production (camembert cheese and white mould cheeses in general) often declared by the use of invalid synonyms as P. album, P. candidum, P. casei or P. caseicola. It is also found as a spontaneous coloniser on fermented sausages originating from the local mycobiota of the production plant (Sunesen 2003) and as a starter culture to give aroma to fermented meat products. This species is also used for enzyme production (Pariza and Johnson 2001). The taxonomy of P. camemberti is well known and this species is accepted as a domesticated form of P. commune. 

There are no reports of an adverse health effect for cheese or meat produced with P. camemberti, i.e. no acute toxicity associated with food produced by P. camemberti has been reported. This species is however known as a producer of cyclopiazonic acid (CPA), this being a neurotoxic and immunosuppressive compound (Frisvad 2004); unknown cytotoxic metabolites are also produced when this fungus is used as a starter culture for mould-ripened meat products (Gareis 1999). A few strains also produce metabolites with poorly described biological activity, such as cyclopaldic acid, rugulovasine A & B and palitantin (Frisvad 2004). CPA has been detected in cheeses at 0.25-0.37 mg/kg cited by (Pitt 1997) and in meat products cited by (Sunesen 2003). Naturally occurring mutants that do not produce this mycotoxin have been reported (Geisen, Glenn et al. 1990). There are not enough toxicological data available to set a threshold under which the consumption of cyclopiazonic acid does not pose any risk. It is important to note that P. camemberti (mostly cited as P. casei) is known as the aetiological agent of the “cheese worker’s lung“ associated with hypersensitivity pneumonitis (Campbell, Kryda et al. 1983; Marcer, Franchini et al. 1996). 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 6 

Despite the long history of apparent safe use of P. camemberti, the capacity of this microorganism to produce cyclopiazonic acid, even under known production conditions, makes P. camemberti ineligible for QPS status. 

2.2. Penicillium chrysogenum 

Penicillium chrysogenum is used as a starter culture for the production of dry sausages (Sunesen 2003) and is also used in the pharmaceutical industry to produce penicillin. It is also known to produce roquefortine C, PR-toxin and secalonic acids, which are mycotoxins, in addition to secondary metabolites with poorly described biological activity: chrysogine, xanthocillin, sorrentanone and sorbicillin (Frisvad 2004). 

Considering the capacity of this species to produce unwanted antibiotics in food, each strain should be investigated in detail, which makes P. chrysogenum ineligible for QPS status. 

2.3. Penicillium funiculosum 

4. funiculosum is used as a producer of enzyme preparation intended for animal feed (Das and Singh 2004) and also as a host for the production of heterologous proteins. The promoter of the histone H4.1 gene was successfully used to drive the expression of an intracellular bacterial enzyme, β-glucuronidase, and a secreted homologous enzyme, xylanase C (Belshaw, Haigh et al. 2002). In general, no known mycotoxins are known from this species; however a single strain has been shown to produce the mycotoxin secalonic acid (JC Frisvad, pers. comm.). Cultures of P. funiculosum do produce many secondary metabolites of unknown structure and unknown biological activity, hence P. funiculosum is ineligible for QPS status. 

2.4. Penicillium nalgiovense 

Penicillium nalgiovense is widely used as starter culture for the production of dry sausages (Sunesen 2003). Wild-type isolates from meats and cheeses have green conidia, whereas starter cultures have white conidia. This species produces penicillin and a broad range of secondary metabolites with poorly described biological activity: nalgiovensin, nalgiolaxin, diaporthins and dipodazin (Frisvad 2004). Typically isolates from meats are good producers of penicillin, while cheese isolates produce penicillin in low amounts. In addition, some strains have been found to produce cytotoxic metabolites on nutrient agar and mould-ripened salamis (Gareis 1999). 

Despite the long history of apparent safe use of P. nalgiovense, the capacity of this species to produce unwanted antibiotics and cytotoxic metabolites in food makes P. nalgiovense ineligible for QPS status. 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 7 

2.5. Penicillium roqueforti 

Penicillium roqueforti has a long history of apparent safe use in the production of blue-moulded cheeses, but is also often isolated from rye bread, silage and other acid preserved products. P. roqueforti has also been reported as a source for enzymes used in food processing (Pariza and Johnson 2001). 

Ten years ago, two closely related species, P. paneum and P. carneum, were discovered (Boysen, Skouboe et al. 1996). All three share many ecological and morphological features, which makes it difficult to interpret older literature, however their profiles of secondary metabolites are distinct (Frisvad 2004; Nielsen 2006). 

1. roqueforti sensu stricto produces the mycotoxins roquefortine C & D, PR-toxin, mycophenolic acid, isofumigaclavine A & B and metabolites with poorly described biological activity: citreoisocoumarin and α-amino butyric acid peptides (peptaibols) (Frisvad 2004). The related species P. carneum produces the mycotoxins mycophenolic acid, patulin, roquefortine C, penitrem A, isofumigaclavine A, as well as cyclopaldic acid with a poorly described activity. P. paneum produces the mycotoxins patulin, roquefortine C & D, botryodiploidin and metabolites with poorly described biological activity: marcfortines and citreoisocoumarin (Frisvad 2004). For P. roqueforti sensu stricto roquefortine and PR-toxin productions are occurring in cheese but at amounts that are not considered as toxic for humans (Pitt 1997). There is no data on possible long-term toxic effects. In general toxicological data for P. roqueforti metabolites are insufficient to set a threshold for regulatory purposes. 

Despite the long history of apparent safe use of P. roqueforti, this species is ineligible for QPS status as no validated analytical methods for the mycotoxins exist to qualify for the absence of mycotoxins under production conditions. 

3. TRICHODERMA 

There have been many developments within the taxonomy and systematics of this genus lasting recent years (Druzhinina and Kubicek 2005; Samuels 2006) and interactive identification key to the more than 90 species of Trichoderma and its teleomorph, Hypocrea has been developed based on molecular methods (Druzhinina, Kopchinskiy et al. 2005; Kopchinskiy, Komon et al. 2005) located at http://www.isth.info/index.php Another interactive key based on morphological and cultural characters for identification of Trichoderma and some of its teleomorphs is also available and has many illustrations. This key is located at http://nt.ars-grin.gov/taxadescriptions/keys/TrichodermaIndex.cfm 

The available literature on bioactive compounds from Trichoderma species is extensive and was reviewed some years ago (Sivasithamparam 1998). Since then, numerous reports have published, however no production of compounds classified as mycotoxins have been reported. For many years the production of trichothecene mycotoxins have been associated with several Trichoderma species, but it has now been clarified that the trichothecene producing species is a newly described species, T. brevicompactum, and not any of those species listed below (Nielsen, Grafenhan et al. 2005). 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 8 

Trichoderma species are known to be aggressive and are used as biocontrol agents, however the difficult systematics is a challenge when it comes to identifying exactly which species is involved (Kullnig 2001; Hermosa, Keck et al. 2004). An EU sponsored initiative to evaluate biological control agents, REBECA, has been launched – see http://www.rebeca-net.de for details. 

3.1. Trichoderma reesei 

Trichoderma reesei is widely used for enzyme production and the toxicological evaluations that need to be taken into consideration have been reported (Blumenthal 2004). However, the potential production of trichothecenes can be neglected as this species cannot produce these mycotoxins (Nielsen, Grafenhan et al. 2005). T. reesei is reported to produce peptaibol compounds which are known to disintegrate cell membranes, causing therefore apoptosis (Bruckner and Graf 1983), as well as other biological active cyclopeptides (Sun, Tian et al. 2006). 

Considering the capacity of this species to produce unwanted biological active compounds, each strain should be investigated in detail, which makes T. reesei ineligible for QPS status. 

3.2. Trichoderma harzianum 

Trichoderma harzianum is mainly used as a bio-control agent (Harman, Howell et al. 2004) and some strains are known to be very aggressive to (plant pathogenic) mushrooms (Samuels 2002). T. harzianum is known to produce a high number of secondary metabolites with partly characterised biological activity (Sivasithamparam 1998; Hanson 2005); however it is known that 6-n-pentyl-α- pyrone (coconut smell) is responsible for at least part of the biological aggressiveness of this species and that highly biologically active α-amino butyric acid cyclic peptides (peptaibols) are involved in the apoptosis mechanism, in addition to anthraquinones, azaphilones, harzianolide and harzianopyrione which have different activities towards plant pathogens (Vinale, Marra et al. 2006). 

Considering the capacity of this species to produce unwanted biological active compounds, each strain should be investigated in detail, which makes T. harzianum ineligible for QPS status. 

3.3. Trichoderma viride 

Trichoderma viride has been evaluated for single cell production (Hang 1976; Youssef 1999), but this has never been commercialised. This species is not used as a bio-control agent but is considered very aggressive and has been reported to produce 6-n-pentyl-α-pyrone (coconut smell) and several biologically active α-amino butyric acid cyclic peptides (peptaibols) in addition to many secondary metabolites with poorly described biological activity (Sivasithamparam 1998). Possibly many production strains are misidentified according to an updated taxonomy; however this cannot be proven as many strains are no longer available. T. viride has been associated with human fungal infections (De Miguel, Gomez et al. 2005). 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 9 

Considering the capacity of this species to produce many biological active compounds, each strain should be investigated in detail, which makes T. viride ineligible for QPS status. 

3.4. Trichoderma longibrachiatum 

Trichoderma longibrachiatum has been reported as a potential bio-control agent (Kullnig 2001; Vizcaino, Sanz et al. 2005). This species is considered very aggressive and has been reported to produce several biologically active α-amino butyric acid cyclic peptides (peptaibols) (Mohamed- Benkada, Montagu et al. 2006) in addition to many secondary metabolites with poorly described biological activity (Sivasithamparam 1998; Sperry 1998; Vicente, Cabello et al. 2001). Possibly many production strains are misidentified according to an updated taxonomy; however this cannot be proven as many strains are no longer available. T. longibrachiatum has been associated with from human fungal infections (Kuhls, Lieckfeldt et al. 1999; De Miguel, Gomez et al. 2005). 

Considering the capacity of this species to produce many biological active compounds, each strain should be investigated in detail, which makes T. longibrachiatum ineligible for QPS status. 

4. FUSARIUM 

Currently, the genus Fusarium contains about 150 species; however the systematics are now changing rapidly due to the rapid developments in molecular biology. Many recently-described Fusarium species have been discovered by molecular tools used in phylogenetic studies, followed by a formal description of the species (Skovgaard 2003; O'Donnell, Ward et al. 2004; Aoki 2005). Introductions to Fusarium are available (Leslie 2001; Summerell 2003; Samson 2004; Leslie 2006) along with extended information on the mycotoxin production by Fusarium species (Marasas 1984; Thrane 2001; Sewram, Mshicileli et al. 2005; Andersen and Thrane 2006). Only one species is used in food production. 

4.1. Fusarium venenatum 

The only commercial mycoprotein products for human food are based on Fusarium venenatum biomass (Quorn® products from Marlow Foods Ltd.). The biotechnological development of these products is well described (Wiebe 2002). The major concern is that F. venenatum is a potential producer of mycotoxins, such as trichothecenes (diacetoxyscirpenol [DAS]) and several derivatives thereof, nivalenol and fusarenon X), butenolide and culmorin (Thrane and Hansen 1995; Miller 2000; Nielsen and Thrane 2001), which are carefully controlled and monitored during mycoprotein production (Johnstone 1998). Strains of F. venenatum which are used to produce enzymes are genetically modified (Royer, Moyer et al. 1995; Royer, Christianson et al. 1999; Pedersen and Broadmeadow 2000; Ahmad, Brinch et al. 2004). 

Considering that F. venenatum is very toxic as wild-type, and that all strains used for enzyme production are genetically modified, this species is ineligible for QPS status. 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 10 

5. MONASCUS 

Monascus species (M. purpureus, M. ruber, M. spp) are known to produce yellow, orange and red pigments. Traditionally, Monascus has been cultured on rice and other cereals by solid sate fermentation. The red-coloured rice (Anka or Ang-kak) has been used for centuries in Asia as natural food colorant for bean curd, meat, wine and other foods. Nowadays, the purified pigments are widely used as colorants in processed seafood, sausages and sauce in Asia. In addition, extracts and other red-mould rice preparations are sold through the internet as nutritional additives with claims that they will lower blood cholesterol levels. No direct adverse health aspects have been reported. However, several studies have shown the presence of the mycotoxin citrinin, which is nephrotoxic and therefore an undesirable toxic secondary metabolite, among the pigments of Monascus and in commercial Monascus-preparations (Blanc, Laussac et al. 1995; Dietrich 1999; Xu 1999). The allergenic relevance of M. purpureus was the first time shown in 2002 (Hipler, Wigger-Alberti et al. 2002). 

The European Community legislation on food additives is based on the principle that only those additives that are explicitly authorised may be used. Pigments from Monascus and Monascus preparations are not included in the list of permitted food colours of the European Parliament and Council Directive. As no toxicological safe and technologically efficient strains of Monascus are available for general use, no species within this genus is eligible for QPS status. 

6. RHIZOMUCOR 

Rhizomucor miehei and Rh. pusillus are the valid names for the thermophilic fungi Mucor miehei and M. pusillus, respectively (Schipper 1978) and both are used to produce chymosin, dextranase and protease with rennet-like activity (Pariza and Johnson 2001). Extensive literature searches have not retrieved any information on toxic compounds produced by these two species. In addition, the WHO has evaluated enzymatic preparations from Rh. miehei and Rh. pusillus and concluded that no adverse effects could be observed (WHO 1975; WHO 1975). 

Despite the apparent safe use as an enzyme producing organism, it has not been possible through extensive literature searches to verify a general absence of biological active secondary metabolites, including allergenic compounds, from Rhizomucor species. Thus, species from this genus cannot be proposed for QPS status. 

7. CRYPHONECTRIA PARASITICA (SYN. ENDOTHIA PARASITICA) 

Cryphonetria parasitica is the valid name for Endothia parasitica (Barr 1978) and this fungus is used to produce protease with rennet-like activity (Pariza and Johnson 2001). WHO has evaluated enzymatic preparations from Cr. parasitica and concluded that no adverse effects could be observed (WHO 1975). However, this species have been reported to produce rugulosin and skyrin (Frisvad 1993). These compounds with poorly described biological activity have also been found in the fermentation batches. 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 11 

Despite the long history of apparent safe use of enzyme production by Cryphonetria parasitica, the capacity of this microorganism to produce biological active compounds under production conditions makes it ineligible for QPS status. 

8. BLAKESLEA TRISPORA 

Blakeslea trispora is used to produce carotenoids in well established commercial products; these naturally produced food colorants are usually not purified. An extensive literature search did not reveal any information on toxic metabolites from this species. In addition, the AFC Panel of EFSA concluded that the toxicity data on lycopene from B. trispora is not of concern as long as the mean intake of lycopene from coloured food does not exceed the intake from natural sources (EFSA 2005). 

Despite the apparent safe use as a colorant producing organism, it has not been possible through extensive literature searches to verify a general absence of biological active secondary metabolites, including allergenic compounds, from Blakeslea trispora. Thus, this species cannot be proposed for QPS status. 

CONCLUSION 

No filamentous fungi can be proposed for a QPS status. The rationale for this is that the methods for identification of fungal cultures to genus/species level are very difficult and often need mycological expertise. There is an ongoing debate on species concepts in the mycological society which result in a lack of a universally accepted fungal taxonomy. This makes identification of fungal cultures intended for commercial use not a trivial issue and often the result should be verified by one or more independent specialists. For the time being there is no universally accepted method for fungal identification. 

The body of knowledge concerning production of toxic compounds is insufficient, as far too little is known about the factors controlling the production of these compounds. In several cases it has been demonstrated that toxic compounds can be produced under production conditions, but often this information is not available. In addition, there are only few validated and certified analytical methods for the detection of a limited number of mycotoxins. For the majority of fungal secondary metabolites no validated method exists. 

The body of knowledge concerning the toxicology of fungal secondary metabolites is insufficient. Bioassays are developed to address specific needs and are not validated. Often the toxicological knowledge is of little or no relevance to real life situations, e.g. lack of information on synergistic effects. The long history of use is not equal to safety, as many fungal metabolites are known to affect the immune system, which could lead to secondary infections. Also the knowledge on long-term effects is insufficient. In conclusion, all fungal species and strains notified to EFSA should be evaluated on a case-by-case basis. 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 12 

REFERENCES 

Abarca, M. L., F. Accensi, et al. (2004). "Taxonomy and significance of black aspergilli." Antonie Van Leeuwenhoek 86(1): 33-49. 

Abarca, M. L., M. R. Bragulat, et al. (1994). "Ochratoxin A production by strains of Aspergillus niger var. niger." Appl Environ Microbiol 60(7): 2650-2. 

Ahmad, S. K., D. S. Brinch, et al. (2004). "Toxicological studies on Lactose Oxidase from Microdochium nivale expressed in Fusarium venenatum." Regul Toxicol Pharmacol 39(3): 256-70. 

Andersen, B. and U. Thrane (2006). "Food-borne fungi in fruit and cereals and their production of mycotoxins." Adv Exp Med Biol 571: 137-52. 

Aoki, T., O'Donnell, K., Scandiani, M.M. (2005). "Sudden death syndrome of soybean in South America is caused by four species of Fusarium: Fusarium brasiliense sp. nov., F. cuneirostrum sp. nov., F. tucumaniae, and F. virguliforme." Mycoscience 46: 162-183. 

Archer, D. B. (2000). "Filamentous fungi as microbial cell factories for food use." Curr Opin Biotechnol 11(5): 478-83. 

Barr, M. E. (1978). "The Diaporthales in North America, with emphasis on Gnomonia and its segregates." Mycologia Memoir 7: 1-232. 

Belshaw, N. J., N. P. Haigh, et al. (2002). "Use of a histone H4 promoter to drive the expression of homologous and heterologous proteins by Penicillium funiculosum." Appl Microbiol Biotechnol 60(4): 455-60. 

Blanc, P. J., J. P. Laussac, et al. (1995). "Characterization of monascidin A from Monascus as citrinin." Int J Food Microbiol 27(2-3): 201-13. 

Blumenthal, C. Z. (2004). "Production of toxic metabolites in Aspergillus niger, Aspergillus oryzae, and Trichoderma reesei: justification of mycotoxin testing in food grade enzyme preparations derived from the three fungi." Regul Toxicol Pharmacol 39(2): 214-28. 

Boysen, M., P. Skouboe, et al. (1996). "Reclassification of the Penicillium roqueforti group into three species on the basis of molecular genetic and biochemical profiles." Microbiology 142 (Pt 3): 541-9. 

Bragulat, M. R., M. L. Abarca, et al. (2001). "An easy screening method for fungi producing ochratoxin A in pure culture." Int J Food Microbiol 71(2-3): 139-44. 

Bruckner, H. and H. Graf (1983). "Paracelsin, a peptide antibiotic containing alpha-aminoisobutyric acid, isolated from Trichoderma reesei Simmons. Part A." Experientia 39(5): 528-30. 

Campbell, J. A., M. J. Kryda, et al. (1983). "Cheese worker's hypersensitivity pneumonitis." Am Rev Respir Dis 127(4): 495-6. 

Cary, J. W. and K. C. Ehrlich (2006). "Aflatoxigenicity in Aspergillus: molecular genetics, phylogenetic relationships and evolutionary implications." Mycopathologia 162(3): 167-77. 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 13 

CAST (2003). Mycotoxins: Risk in Plant, Animal and Human Systems. Task Force Report n°139. Ames (Iowa, USA), Council for Agricultural Science and Technology - CAST. 

Chang, P. K., K. C. Ehrlich, et al. (2006). "Cladal relatedness among Aspergillus oryzae isolates and Aspergillus flavus S and L morphotype isolates." Int J Food Microbiol 108(2): 172-7. 

Chattopadhyay, S. K., B. Nandi, et al. (1987). "A new mycotoxin from Aspergillus candidus Link isolated from rough rice." Mycopathologia 98(1): 21-6. 

Christias, C., C. Couvaraki, et al. (1975). "Protein content and amino acid composition of certain fungi evaluated for microbial protein production." Appl Microbiol 29(2): 250-4. 

Das, H. and S. K. Singh (2004). "Useful byproducts from cellulosic wastes of agriculture and food industry--a critical appraisal." Crit Rev Food Sci Nutr 44(2): 77-89. 

De Miguel, D., P. Gomez, et al. (2005). "Nonfatal pulmonary Trichoderma viride infection in an adult patient with acute myeloid leukemia: report of one case and review of the literature." Diagn Microbiol Infect Dis 53(1): 33-7. 

Dietrich, R., Usleber, E., Martlbauer, E., Gareis, M. (1999). "Detection of the nephrotoxic mycotoxin citrinin in foods and food colorants derived from Monascus spp." Archiv fur Lebensmittelhygiene 50: 17-21. 

Druzhinina, I. and C. P. Kubicek (2005). "Species concepts and biodiversity in Trichoderma and Hypocrea: from aggregate species to species clusters?" J Zhejiang Univ Sci B 6(2): 100-12. 

Druzhinina, I. S., A. G. Kopchinskiy, et al. (2005). "An oligonucleotide barcode for species identification in Trichoderma and Hypocrea." Fungal Genet Biol 42(10): 813-28. 

EFSA (2005). "Opinion of the Scientific Panel on Food Additives, Flavourings, Processing Aids and Materials in Contact with Food on a request from the Commission related to an application on the use of a-tocopherol containing oil suspensions and cold water dispersible forms of lycopene from Blakeslea trispora as a food colour." The EFSA Journal 275: 1-17. 

EFSA (2006). "Opinion of the Scientific Panel on Additives and Products or Substances used in Animal Feed on safety and efficacy of the product "Amaferm" as a feed additive for dairy cows and cattle for fattening in accordance with Regulation (EC) No 1831/2003." The EFSA Journal 337: 1- 17. 

Frisvad, J. C., K. F. Nielsen, et al. (2006). "Recommendations concerning the chronic problem of misidentification of mycotoxigenic fungi associated with foods and feeds." Adv Exp Med Biol 571: 33-46. 

Frisvad, J. C., Samson, R.A. (2004). "Polyphasic taxonomy of Penicillium subgenus Penicillium. A guide to identification of food and air-borne terverticillate Penicillia and their mycotoxins." Studies in Mycology 49: 1-173. 

Frisvad, J. C., Smedsgaard, J., Larsen, T.O., Samson, R.A. (2004). "Mycotoxins, drugs and other extrolites produced by species in Penicillium subgenus Penicillium." Studies in Mycology 49: 201- 242. 

Appendix D - Assessment of filamentous fungi 

The EFSA Journal (2007) 587, Qualified Presumption of Safety 14 

All statements herein have not been evaluated by the Food and Drug Administration. The products on this website were not intended to diagnose, treat, cure, or prevent any disease.