How common is it for environmentally occurring bacillus strain to contain some sort of plasmid?

How common is it for environmentally occurring bacillus strain to contain some sort of plasmid?

We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

I am designing a method for testing whether two new bacillus isolates, that are ionic silver resistant, store the silver resistance mechanism on a plasmid or on the chromosome of the bacteria.

In my method, after I do a plasmid extraction on the bacteria, I want to use nanodrop to see if plasmids are even present in the bacteria (if no plasmids are found, then the silver resistance mechanism must be stored chromosomally). After this, I want to use heat shock or electroporation to place the plasmid in a silver sensitive bacteria, and see if the newly transformed bacteria is ionic silver resistant (this would suggest the resistance mechanism is stored on the plasmid)

I have been looking at how other people have tested for this, and I have found that the nanodrop part is not done. I thought the nanodrop would be a good control to make sure the plasmid extraction went well, and if plasmid DNA was not found, it would save me from doing the rest of the experiment.

One reason why I thought they didn't do nanodrop was that perhaps most environmentally occurring bacillus contain plasmids, and thus nanodrop would be unnecessary step.

So my question becomes how common is it for environmentally occurring bacillus strain to contain some sort of plasmid? Also, if anybody has any suggestions for improvment of my method, please let me know.

Plasmids are widespread among bacteria and are important because they spread virulence and antibiotic resistance traits, among others. They are horizontally transferred between strains and species, so it is difficult to work out their evolution and epidemiology. Agrobacteria, a diverse grouping of species that infect plants, inject oncogenic Ti and Ri plasmids, which cause crown galls and hairy root diseases, respectively. The upside is that these plasmids have become valuable biotechnological tools. Weisberg et al. combed through an 80-year-old collection of Agrobacterium strains but found a surprisingly low diversity of plasmids. It is puzzling how limited the number of plasmid lineages is despite reported high levels of plasmid recombination, but what is clear is how plant production systems have influenced plasmid spread into various genomic backbones.


Plasmids are autonomously replicating, nonessential DNA molecules that accelerate the evolution of many important bacterial-driven processes. For example, plasmids spread antibiotic resistance genes, which are a pressing problem for human and animal health. Plasmids can also encode complex traits that allow bacteria to interact intimately with eukaryotes. Acquisition of an oncogenic tumor-inducing (Ti) or root-inducing (Ri) plasmid by saprophytic soil agrobacteria changes them into pathogens capable of genetically transforming and causing disease in a broad range of plant species.

Plasmids are also biotechnology tools that can advance our understanding of life. They can be used to generate organisms with unusual traits and innovative applications. The potential for using oncogenic plasmids to accelerate research was recognized early in their discovery. Along with strains of agrobacteria, disarmed plasmids are mainstays as tools in plant biology and plant genetic engineering.


Inferring evolutionary relationships is foundational for classifying plasmids, accurately assessing the influence of plasmids on disease outbreaks, developing appropriate strategies for mitigating disease, and expediting efforts to leverage plasmid diversity for biotechnology. However, such research is complicated because plasmids consist of diverse structural variants and are extraordinarily dynamic, modular molecules that can be reshuffled and broadly transmitted horizontally.

We focused on oncogenic plasmids of agrobacteria because of their important roles in causing disease and as biotechnology tools. Two genomic datasets were developed. One consisted of diverse, broadly sampled historical strains and was intended to serve as the basis for an evolutionary framework. The other consisted of contemporaneous strains hierarchically sampled from managed plant production sites, for the purpose of calibrating epidemiology methods. The datasets were combined to identify epidemiological patterns.


We combined analyses of chromosomal ancestry and plasmids to uncover their contributions and accurately model the global spread of disease. Phylogenetic, genomic, and time tree analyses of thousands of strains from the Rhizobiales order yielded a robust phylogenetic history of agrobacteria. We developed a strategy that uses phylogenetic and network approaches as well as different scales of genetic information to infer the evolution of diverse oncogenic plasmids. By combining results, we uncovered global epidemiological patterns supporting movement of pathogens clonally and plasmids horizontally in space and time.

This study has three major findings: (i) Lineages of agrobacteria emerged independently and at different times from within a genus-level group that also circumscribes multiple lineages of rhizobia. (ii) Agrobacterial Ti and Ri plasmids are descended from only six and three lineages (types), respectively. Few evolutionary events are sufficient to explain the relationships observed among types. Each type is subject to different pressures and shows different degrees of within-group variation, but their evolution is nonetheless guided by similar principles. The extent of modularity is high, and genes and functional modules are frequently reshuffled via recombination within conserved loci. Yet plasmid diversification is nonetheless constrained by the spatial structure of loci that interact genetically. (iii) Transmission of oncogenic plasmids, especially within agricultural settings, promotes the massive spread of disease.


Our strategy for inferring the evolution and transmission of virulence plasmids has potential applications in plant and human or animal health and food safety, as well as for understanding the ecology and evolution of other plasmid-mediated processes such as mutualistic symbioses. In addition, this strategy can be applied to study other mobile and modular elements, such as integrative conjugative elements and pathogenicity or symbiosis islands. We have shown that plasmids once viewed as too diverse to be classified have distinct lineages, and that accurate modeling of the spread of disease can be accomplished by robustly defining their evolutionary relationships.

Genomic data from hundreds of strains of agrobacteria were parsed and analyzed to infer the evolutionary histories of chromosomes and oncogenic Ti and Ri plasmids. The data were overlaid to uncover the roles of bacteria and plasmids in the global spread of disease.

Conjugating plasmids into bacteria

The helper plasmid, based on the naturally occurring pRK plasmid contains the genes necessary for encoding conjugation machinery. The machinery is promiscuous. It can transfer origin of transfer (oriT) regions from different plasmid groups.

The donor should contains an oriT. (temperature sensitive origins, or pEX18 plasmids should be kept at appropriate temperatures)

  • Inoculate (i) helper: into Kan plate(ii) the donor plasmid (into the appropriate antibiotic plate) and the recipient (Salmonella, Pseudomonas…etc).
  • Next day, scrape about a loopful bacteria (about a pepper corn size (no agar pieces).
  • Resuspend in 500ul LB broth, wash once. Spin gently to avoid sex pili breakage.
  • Resuspend in 100ul LB. Mix all together. Spot 100ul into a very dry LB agar plate. Incubate at 37C or 30C.
  • (for mating in deletion plasmids, once three bacteria are mixed, spin down gently, resuspend in 100ul)

No need to use filters for mating.

If you are conjugating a replicative plasmid, after 4-6 hours streak for single colonies in a selective agar media. If moving a suicide vector, grow overnight. Next day, streak for single colonies on selection plates. The selection plate should kill both the helper and the donor E. coli but not the recipient.

Selection plates:

For Pseudomonas: Use Irgasan (Triclosan) at 25ug/ml final into LB agar containing the appropriate donor selectable antibiotic. Dissolve the powder in 100% alcohol for stocks.

E. coli K12 cannot transport glutamate and citrate. Most lab-derived E. coli strain used for molecular biology have some kind of auxotrophy (thi, leu, pro). So they will die in a media lacking a particular nutrient. thi mutation, encoding functions for making Vit B1, is common in lab E. coli strains (DH5alpha, SM10lambdaPir, MC1061lambdaPir)

  • Autoclave 850ml water and 15g agar, with a stirbar.
  • After cooling down, add the following.

This media is also effective for Salmonella SL14028. (use Iron at 10uM).
VBMM media can also be used for conjugant selection.

Diparental mating: Difficult conjugations can be performed by transforming the donor plasmid into E. coli SM10 strain which contains the pRK conjugation apparatus integrated into its chromosome. SM10 is KanR.

Certain P. aeruginosa (eg PAO1), Salmonella and Klebsiella strains presumably contain restriction systems that will severely restrict foreign DNA. In that case, incubate the recipient at 42C for 2 hours to overnight before conjugation.


Susceptible insects must ingest Bt toxin crystals in order to be affected. In contrast to poisonous insecticides that target the nervous system, Bt acts by producing a protein that blocks the digestive system of the insect, effectively starving it. Bt is a fast-acting insecticide: an infected insect will stop feeding within hours of ingestion and will die, generally from starvation or a rupture of the digestive system, within days.

Whether applied in spray form or through genetic engineering, each Bt strain is effective against a narrow range of insects. The most commonly used strain of Bt (kurstaki, or Btk) targets only certain species of caterpillars. Since the late 1970s, Bt strains (e.g., israelensis, or Bti) have been developed that control certain types of fly larvae, including those of mosquitoes, black flies, and fungus gnats. Other common strains include san diego and tenebrionis, which are effective against certain leaf beetles, such as the Colorado potato beetle and elm leaf beetle.

Bacillus thuringiensis (Bt)

Bt is a microbe naturally found in soil. It makes proteins that are toxic to immature insects (larvae). There are many types of Bt. Each targets different insect groups. Target insects include beetles, mosquitoes, black flies, caterpillars, and moths.

With Bt pesticides, routine testing is required to ensure that unwanted toxins and microbes are not present. Bt has been registered for use in pesticides by the US Environmental Protection Agency (EPA) since 1961.

What are some products that contain Bacillus thuringiensis (Bt)?

Currently, Bt strains are found in over 180 registered pesticide products. Bt products are used on crops and ornamental plants. Others are used in and around buildings, in aquatic settings, and in aerial applications. These products are commonly sprays, dusts, granules, and pellets. Some of these products are approved for use in organic agriculture.

Some crops have been engineered to make the Bt toxin. These plant-incorporated protectants include corn, cotton, and soybeans.

Always follow label instructions and take steps to avoid exposure. If any exposures occur, be sure to follow the First Aid instructions on the product label carefully. For additional treatment advice, contact the Poison Control Center at 1-800-222-1222. If you wish to discuss a pesticide problem, please call 1-800-858-7378.

How does Bacillus thuringiensis (Bt) work?

Bt makes toxins that target insect larvae when eaten. In their gut, the toxins are activated. The activated toxin breaks down their gut, and the insects die of infection and starvation. Death can occur within a few hours or weeks.

The different types of Bt create toxins that can only be activated by the target insect larvae. In contrast, when people eat the same toxins, the toxins are not activated and no harm occurs.

Each type of Bt toxin is highly specific to the target insect. For example, the ‘kurstaki’ type targets caterpillars. The ‘isrealensis’ type targets immature flies and mosquitoes. Little to no direct toxicity to non-target insects has been observed.

How might I be exposed to Bacillus thuringiensis (Bt)?

People are most commonly exposed to Bt through their diet, at very low levels. Exposure can also occur if you breathe it in or get it on your skin or eyes. For example, this can occur while applying sprays or dusts during windy conditions. You may also be exposed after using a product if you don’t wash your hands before eating or smoking. Since Bt is commonly found in soils, exposures not related to pesticides are also possible.

Pets might be exposed to this product in treated birdbaths or water fountains. You can limit your exposure and reduce the risk by carefully following the label instructions.

What are some signs and symptoms from a brief exposure to Bacillus thuringiensis (Bt)?

Bt is low in toxicity to people and other mammals. Several studies have found no evidence of sickness or infection as a result of exposure. However, some products with Bt have caused eye and skin irritation. In one study, rats breathed in very high doses of concentrated Bt. Some had runny noses, crusty eyes, and goose bumps. Others were less active or lost weight.

In another study, people were surveyed before and after aerial applications of Bt. Most people were not affected. However, some people with hay fever reported certain symptoms. These included difficulty with sleep and concentration, stomach upset, and nose/throat irritation. Seasonal factors, such as pollen, may have contributed to some of the effects.

Scientists also evaluated whether Bt can cause allergic reactions. Researchers found that farmworkers exposed for one to four months did not experience any problems related to their airways, nose, or skin. However, further exposure showed evidence of an immune response and the potential for skin allergies to develop.

What happens to Bacillus thuringiensis (Bt) when it enters the body?

When eaten, Bt is confined to the gut. It does not reproduce, and the toxin is broken down like other proteins in the diet. Bt leaves the body within 2 to 3 days.

If breathed in, Bt can move to the lungs, blood, lymph, and kidneys. Bt is then attacked by the immune system. Levels of Bt decrease quickly one day after exposure.

Is Bacillus thuringiensis (Bt) likely to contribute to the development of cancer?

No data were found on the carcinogenic effects of Bt in humans. However, in one animal study, rats were fed very high doses of Bt for 2 years. No evidence of cancer was observed.

Has anyone studied non-cancer effects from long-term exposure to Bacillus thuringiensis (Bt)?

In a 2-year study, rats were fed high doses of Bt daily. Female rats had lower body weights. However, no evidence of an infection was found.

Bt is only activated in the alkaline environment of the insect gut, compared to the acidic environment of human stomachs. In human stomachs, it is easily digested. As such, no adverse effects are expected after long-term dietary exposure to Bt, whether its proteins are sprayed on plants or grown within plant tissues.

Are children more sensitive to Bacillus thuringiensis (Bt) than adults?

Children may be especially sensitive to pesticides compared to adults. However, there are currently no data showing that children have increased sensitivity specifically to Bt.

What happens to Bacillus thuringiensis (Bt) in the environment?

Toxins created by Bt are rapidly broken down by sunlight and in acidic soil. Other microbes in soil can also break it down. Bt does not readily leach in soil. It typically remains in the top several inches of soil. Bt remains dormant in most natural soil conditions. However, there has been some reproduction in nutrient rich soils. On the soil surface, dormant Bt cells last only a few days. However, below the soil surface, they can last for months or years. The half-life in unfavorable soil is about 4 months. Bt toxins break down much faster. In one study, 12% remained after 15 days.

In water, Bt does not readily reproduce. A study found Bt toxins in the air were broken down rapidly by sunlight. Forty-one percent (41%) of the toxin remained after 24 hours. On plant surfaces, sunlight breaks down Bt the half-life of Bt toxins is 1-4 days.

Can Bacillus thuringiensis (Bt) affect birds, fish, or other wildlife?

Bt is practically non-toxic and non-pathogenic to birds, fish, and shrimp. No adverse effect or infection was found in rats given large doses of two different Bt strains. There is no evidence that Bt can cause a disease outbreak among wild animals.

Little to no direct toxicity to non-target insects and other shelled invertebrates has been observed. Bt does not seem to hurt earthworms. However, the aizawai strain is highly toxic to honeybees. Other strains have minimal toxicity to honeybees.

Water fleas exposed to the kurstaki and israelensis strains showed moderate toxicity. The aizawai strains are highly toxic to water fleas. However, evidence suggests that toxicity to these non-targets may be related to impurities from the production of Bt.


When a gene from one organism is transferred to improve or induce desired change in another organism, in laboratory, the result is a genetically engineered (or modified) organism (which may also called transgenic organism). There are different methods to transfer genes to animals and plants where the old and most traditional one is through the selective breeding. For example, a plant with a desired trait is selected and bred to produce more plants with such a trait. Recently, with the reached high technology, advanced techniques are carried out in laboratory to transfer genes that express the desired traits from a plant to a new plant (Martineau 2001).

The first produced genetically modified plant in the laboratory was tobacco in 1983 and was tested in 1986 as herbicide-resistant in France and the USA. In 1994, the European Union approved the commercial production of the plant as resistant to the herbicide bromoxynil (Martineau 2001).

Tomato was the first commercially grown genetically modified whole food crop (called FlavrSavr) which was modified to ripen without softening by a Californian company, Calgene (Martineau 2001). Calgene took the initiative to obtain the Food and Drug Administration (FDA) approval for its release in 1994. It was welcomed by consumers who purchased the fruit at high price. However, a conventionally bred variety with longer shelf-life prevented the product from becoming profitable.

World cultivation of commercialized GM crops

In 1997, the total cultivated area of GM crops was 1.7 million ha and increased gradually to reach 185.1 million ha in 26 countries in 2016 19 of these are developing countries beside 7 industrial countries. The GM crops include, mainly, 5 crops: two of them (corn and cotton) are resistant to insects alone or to insects and herbicides together. The other three (soybean, canola and sugar beet) are resistant to herbicides. The area of GM crops in the developing countries in 2016 was 99.6 million ha (54%) while it was 85.5 million ha (46%) in the industrial ones. USA grew 72.9 million ha (representing 39% of the world total area), Brazil (27%), Argentina (13%), Canada (6%), India (6%), Paraguay (2%), Pakistan (2%), China (2%), and South Africa (1%). Five European countries (Spain, Portogal, Czec Republic, Slovakia, and Romania) planted about 117,000 ha in 2015 that increased to 136,000 ha in 2016. Romania decided not to plant in 2016 due to onerous requirement by the government (James 2016).

Genetically engineered products are not new. Insulin used in medicine is an example of genetic engineering. Genes encoding human insulin were cloned and expressed in E. coli in 1978. At present, insulin is being produced in E.coli and the yeast Saccharomyces cerevisiae for diabetic patients (Baeshen et al. 2014).

How to produce genetically modified crops

Rani and Usha (2013) mentioned that the modified crops are produced by:

Identifying and locating genes for plant traits, which is the most limiting step in the transgenic process. Identifying a single gene involved with a trait is not sufficient scientists must understand how the gene is regulated, what other effects is might have on the plant and how it interacts with other genes active in the same biochemical pathway.

Designing genes for insertion, but once a gene has been isolated and cloned (amplified in a bacterial vector), it must undergo several modifications before it can be effectively inserted into a plant.

Transforming plants, which is the heritable change in a cell or organism brought about by the uptake and establishment of introduced DNA. There are two main methods of transforming plant cells and tissues: (a) The gene gun method which has been especially useful in transforming monocot species like corn and rice and (b) the Agrobacterium method which is considered preferable to the gene gun.

Agrobacterium tumefaciens is a soil-dwelling bacterium that has the ability to infect plant cells with a piece of its DNA. When the bacterial DNA is integrated into a plant chromosome, it effectively hijacks the plants’ cellular machinery and uses it to ensure the proliferation of the bacterial population.

Selection of successfully transformed tissues following the gene insertion process to be transferred to a selective medium containing an antibiotic. Only plants expressing the selectable marker gene will survive and possess the transgene of interest.

Regeneration of whole plants under controlled environmental conditions in a series of media containing nutrients and hormones (a process that is known as tissue culture).

This process is performed mainly for the production of insect- or herbicide-resistant crops which are called Genetically Modified Crops (Fiester 2006).

Bt crops

Bt crops are plants genetically engineered (modified) to contain the endospore (or crystal) Bt toxin to be resistant to certain insect pests. “Plant Genetic Systems”, in Belgium, was the first company to produce a Bt crop (tobacco) in laboratory in 1985 but the crop was not commercially successful (Vaek et al. 1987). However, in 1995, the Environmental Protection Agency (EPA) in the USA approved the commercial production and distribution of the Bt crops (corn, cotton, potato, and tobacco). Currently, the most common Bt crops are corn and cotton (Vaek et al. 1987). In 2013, four insect-resistant Bt brinjal (eggplant) varieties were approved for seed production and initial commercialization in Bangladesh (Koch et al. 2015). Recently, Bt soybean varieties expressing Cry1Ac+ Cry1Ab were approved for commercial use in Latin America to control lepidopteran insects (Koch et al. 2015). Bt crops, containing Bt toxins, were planted in almost 100 million ha (Brookes and Barfoot 2017).

The most widely used Bt vegetable crop is sweet corn. Shelton et al. (2013) compared sweet corn varieties grown in the USA where the primary insect pest was Heliothis zea and demonstrated that non-sprayed Bt varieties produced more clean marketable ears than corn varieties sprayed with chemical insecticides up to 8 times.

Adoption of Bt cotton has greatly reduced the abundance of targeted pests in cotton and other crops close to cotton that are infested by polyphagous target insects (Naranjo 2011). In addition, the reduction in insecticide use enabled IPM programs in Bt crops fields and the increase of natural enemies populations.

Bacillus thuringiensis (Bt)

History of Bt

Bt was first discovered in 1901 by the Japanese biologist Shigetane Ishiwatarias a cause of sotto disease that was killing silkworms and named it Bacillus sotto (Milner 1994). In 1911, Ernst Berliner isolated this bacterium from dead Mediterranean flour moth in Thuringia, Germany, and named it Bt. In 1915, Berliner reported the existence of a parasporal body, or crystalline inclusion (called crystal) close to the endospore within Bt spore (Fig. 1), but the activity of the crystal was not then discovered (Milner 1994). In 1956, it was found that the main insecticidal activity against lepidopteran insects was due to the parasporal crystal (Milner 1994). Zakharyan (1979) reported the presence of a plasmid in a strain of Bt and suggested that the plasmids involved in formation of endospore and crystal.

The spore of Bacillus thuringiensis “from Hofte and Whitely (1989)”

In 1938, Bt was commercially produced in France with the name Sporine to be used primarily to kill flour moth (Luthy et al. 1982). In 1956, Bt was used commercially in the USA, but the products were not successful because of poor formulations (Milner 1994). In the 1980s, the use of Bt increased worldwide when insects became increasingly resistant to the chemical insecticides (Milner 1994).

The crystal

The crystal, referred to as Cry toxin (cry from crystal), insecticidal crystal protein, parasporal body, crystalline inclusion, or delta endotoxin, is a protein formed during sporulation in Bt strains and aggregate to form crystals. Such Cry toxins are toxic to specific species of insects belonging to Lepidoptera, Coleoptera, Hymenoptera, Diptera, and Nematoda. They are harmless to human, vertebrates, and natural enemies of insects (Hofte and Whitely 1989).

In addition to the Cry toxins, some strains of Bt, like Bt israelensis, produce another toxic crystal, named cytolytic protein or Cyt toxin. The Cyt toxin (or protein) derived its name from being cytolytic to a wide range of invertebrate and vertebrate cells in vitro. This Cyt toxin increases the efficiency of Bt in dipteran insects (suborder: Nematocera) and some coleopteran ones. The Cyt toxins are also formed during sporulation and occur within the parasporal body but in a separate inclusion. They share no significant amino acid sequence identity with Cry toxins and are thus unrelated (Hofte and Whitely 1989).

Mode of action of Bt

Bt spores have to be ingested by the susceptible insect to cause mortality. The Cry toxin becomes active by proteoletic enzymes in the alkaline gut juice (pH 8–10). Most cry toxins are actually pro-toxins of about 130 to 140 kDa, and after activation, they become 60–70 kDa (Bravo et al. 2007). The activated toxin passes through the peritrophic membrane and binds to specific receptors on apical microvillar brush border membrane of the epithelial cells of the midgut making pores through which the toxin penetrates to such cells that become swollen. The swelling continues until the cells lyse and separate from the basement membrane of the midgut epithelium. The alkaline gut juices then leak into the hemocoel causing the hemolymph pH rises that leads to paralysis and death of the insect (Soberon et al. 2010). However, Broderick et al. (2006) mentioned that the naturally occurring bacteria in the gut (E.coli and Enterobacter) penetrate to the hemocoel through the disrupted epithelium caused by Bt toxins and multiply causing sepsis of the hemolymph and death of the insect. In the Bt-moderately sensitive insects, such as Spodoptera spp., the endospore has a considerable role in killing the insect by producing toxins during its vegetative growth in the hemolymph (Crickmore et al. 2014). The insect or any living organism that does not have the receptors in gut epithelial cells is not killed by Bt (Gill et al. 1992).

The Cyt toxin is also a protoxin, about 28 kDa, and is activated by the proteolytic enzymes in the midgut juice to become 24 kDa. The toxin then penetrates from peritrophic membrane and the epithelial cells which lyse and separate causing the death of the insect (Hofte and Whitely 1989).

Nomenclature of Cry and Cyt toxins (proteins)

In the early 1980s, it was discovered that “there are genes responsible of the production of the crystal proteins in Bt spore and these genes are carried on plasmids” (Crickmore et al. 1998). Hofte and Whitely (1989) termed these “cry genes” and the protein they encode “cry proteins” (for crystal) and cyt proteins (for cytolytic). They classified these genes, or crystals, based on the spectrum of activity of the proteins (insect order), their size or mass, and their apparent relatedness as deduced from nucleotide and amino acid sequences.

This designation was followed by a Roman numeral that indicates patho-type (I and II for toxicity to lepidopterans, III for toxicity to coleopterans and IV for toxicity to dipterans). This numeral was followed by an uppercase letter indicating the chronological order in which genes with significant differences in nucleotide sequences were described.

As the number of Bt Cry and Cyt toxins increased, the nomenclature of Hofte and Whitely (1989) was modified as follows: the Cry and Cyt were maintained but the Roman numerals were replaced with Arabic numbers (Cry1 and Cry2 for toxicity to lepidopterans Cry3 for toxicity to coleopterans Cry4, Cry10, and Cry11 for toxicity to dipterans). The numbers (1, 2, 3,….) indicate major relationships (90% identity). The uppercase letters (A, B, C….) indicate 95% identity. Minor variations were designated by lowercase letters (a, b, c….), for example, Cry1Aa, Cry2Ab, …… (Table 1). So, the structure of Cry1Aa differs slightly from that of Cry1Ab.

For more and recent information about nomenclature of Bt toxins, see Crickmore (2017).

The crystal often contains one or more Cry toxins (or genes). For instance, the crystal of Bt kurstaki contains four cry genes (or toxins) Cry1Aa, Cry1Ab, Cry1Ac, and Cry2Aa (Fig. 2). In contrast, Bt thompsoni contains only one cry toxin: Cry3Aa. The existence of more than Cry toxin in a Bt strain increases the efficiency and the host range of this strain.

Crystal of Bt kurestaki and Bt israelinsis “From Hofte and Whitely (1989)”. a Bt spore. b Crystal of Bt kurestaki. c Crystal of Bt israelensis. The arrows in a illustrating the endospore (on the left) and the crystal (on the right)

The crystal of Bt israelensis, however, contains four Cry toxins (or genes): Cry4Aa, Cry4Ba, Cry10Aa, and Cry11Aa in addition to two Cyt toxins, Cyt1Aa, and Cyt2Ba (Ben-Dov 2014) (Fig. 2). Despite the low toxicity of the two Cyt toxins, they are highly synergistic with the Cry toxins that increase the toxicity of Bt israelensis by 3–5-folds than the Cry proteins alone (Ben-Dov 2014).

Interestingly, Palma et al. (2014) reported that 700 Bt cry genes (Cry proteins) have been identified in the past decades. While many Cry proteins have pesticidal properties against insect pests in agriculture, others have no known invertebrate targets and have been termed “parasporins”. Some of this parasporin group of Bt Cry proteins such as Cry31A, Cry41A, Cry46A and Cry64A exhibit strong and specific cytocidal activity against human cancer cells of various origins. They have been given the alternative names: parasporin-1 (PS1), parasporin-2 (PS2), parasporin-3 (PS3), parasporin-4 (PS4), parasporin-5 (PS5), and parasporin-6 (PS6).

DNA, gene, and plasmid

DNA (deoxyribonucleic acid) is a very large molecule that carries the genetic instructions used in growth, development, functioning, and reproduction of all living organisms and many viruses. It consists of two long nucleotide chains. The nucleotides are composed of a five-carbon sugars to which are attached one or more phosphate group and a nitrogenous base. The sugar is deoxyribose attached to a single phosphate group (representing the backbone of the DNA). The base may be either adenine (A), guanine (G), thiamine (T), and cytosine (C). The nucleotides are linked together in a chain through the sugar and phosphates (Rettner 2017) (Fig. 3).

The structure of DNA (US Natural Library of Medicine)

A gene is a distinct segment of DNA that encodes the information necessary for the assembly of a specific protein. The protein then functions as enzyme to catalyze biochemical reactions, or as a structure or a storage unit of a cell to contribute to expression of a plant trait (Rani and Usha 2013).

A plasmid (Fig. 4) is a small DNA molecule. It naturally exists in bacterial cells and some eukaryotes. Often, the genes carried in plasmids provide the bacteria advantages such as antibiotic resistance (Roh et al. 2007) .

Diagram of the endospore of Bt (from Google)

Vegetative insecticidal proteins (Vip) toxins

In addition to ȣ-indotoxins (Cry and Cyt toxins), Bt produces a novel family of insecticidal proteins named vegetative insecticidal proteins (Vip) during its vegetative stage. Two classes of Vip toxins were described. The first consists of a binary system composed of two proteins: Vip 1 and Vip 2, which are 100 kDa and 52 kDa in size, respectively. These proteins are highly toxic to certain coleopteran species (Chakroun et al. 2016). The second class is of a 88.5 kDa protein (Vip 3) and active against a wide range of lepidopteran insects (Chakroun et al. 2016). These two classes of proteins do not display sequence homology with Cry or Cyt proteins (Chakroun et al. 2016). There are, to date, about 82 identified Vip genes. The Vip toxins do not form crystals.

Currently, available Bt cotton varieties produce either or both Cry toxins and Vip toxins that target specific caterpillar pests such as beet armyworm, Spodoptera exigua cotton bollworm, Helicoverpa armeigera and tobacco budworm, Heliothis virescens.

Bt corn (maize)

Corn is the sole Bt crop commercially produced and sold in 5 European countries (Spain, Portogal, Romania, the Czech Republic, and Slovania) (Koch et al. 2015) and is used for feeding livestock and as row material for the starch industry. Such countries produce approximately 173 million tones ensilage maize and 56 million tons of grain maize. A part of the Bt corn seeds is used for manufacturing food products, like starch, cornflakes, popcorn, canned sweet corn, corn on the cob, and corn oil, as the high heat used for producing such foods breaks down any toxins. There are rules in Europe countries that all food products made from Bt corn must be labeled. The USA and Canada, however, do not have such rules, and almost 75% of their manufactured corn products are made from Bt corn (Anonymous 2012).

Cultivation of Bt corn started in the USA, Canada, and Europe (Spain) in 1997, and by 2009, it was commercially planted in 11 countries. It was then representing 85% of the total area of corn in USA, 84% in Canada, 83% in Argentina, 57% in South Africa, 36% in Brazil, 20% in Spain, and 19% in Philippines (James 2016). In 2016, GM corn in the world (in 16 countries) reached 60.6 million ha, out of which 6 million (10%) were Bt corn, 7 million (11.7%) were herbicide-tolerant corn, and 47.7 million (78.7%) were combined Bt and herbicide-tolerant corn. The crop was produced to resist the infestation by the European corn borer, Ostrinia nubilalis, but later in the 2000s, it has been produced against the corn earworm, H. zea, and the corn rootworm, Diabrotica virgifera in addition to O. nubilalis (James 2016).

Bt cotton

For cotton growers, there was a lot of pressure from pests before the introduction of Bt cotton. Due to synthetic insecticide resistance, farmers were losing much of their cotton because of H. virescens and pink bollworm, Pectinophora gossypiella. According to USDA, 94% of the cotton cultured in USA is genetically modified (James 2016).

A study in University of California revealed that the average cost reduction in pesticides applied in Bt cotton fields from 1996 to 1998 was between 25 and 65 dollars per acre the yield estimated, in the same period, was 5% more, on average, than the traditional cotton. In addition, Bt cotton significantly decreased the number of foliar sprays, against other cotton pests and consequently the cost of insecticides (Anonymous 2000).

In 1996, Bollgard cotton (a trademark of Monsanto Company) was the first Bt cotton to be marketed in the USA. It was producing Cry1Ac toxin with high activity on tobacco budworm and pink bollworm. Bt cotton was widely adopted in the USA by farmers in the Western Cotton Belt for the pink bollworm and by farmers in the Mid-south and South-east for primarily tobacco budworm and to a lesser extent for fall armyworm, Spodoptera frugiperda and S. exigua (Stewart 2007).

Bollgard II was introduced in 2003 representing the next generation of Bt cotton. It was producing Cry2Ab toxin. Wide Strike cotton (a trademark of Dow Agro-sciences) was produced in 2004 containing Cry1Ac and Cry1F. Both Bollgard II and Wide Strike have better activity on a wide range of caterpillar insects than the original Bollgard (Stewart 2007).

The most recent 3rd generation of Bt cotton contained three genes: Bollgard 3 (Cry1Ac + Cry2Ab + Vip3A), Twin Link Plus (Cry1Ab + Cry2Ac + Vip3Aa19), and Wide Strike 3 (Cry1Ac + Cry1F + Vip3A) (Vyavhare 2017).

Bt cotton is the only Bt crop cultivated in developing countries (James 2016). In India and China, the cultivated area of Bt cotton increased sharply during 2006 and 2007 to reach 25 million acres (2.5 million ha). Cultivation of Bt cotton in India started in 2002 (James 2016). In 2016, the world total area of cotton was 35 million ha (in 18 countries), out of which 22.3 million (64%) were GM cotton. In the USA, however, the total area of cotton was 4 million ha and out of which 3.2 million ha (80%) were combined Bt and herbicide-tolerant cotton (James 2016).

Varieties of Bt corn and Bt cotton registered in the USA were producing 18 different combinations of 11 Bt toxins. Each variety produces 1–6 Bt toxins that kill caterpillars, beetles, or both (Tabashnik et al. 2009).

Insects resistance to Bt crops

Insects’ field-evolved resistance is defined as a genetically based decrease in susceptibility of a population to a toxin caused by exposure of the population to the toxin in the field. The main goal of monitoring resistance of insects to Bt crops is to detect resistance early enough to enable taking preventative measures before failures occur (Tabashnik 1994).

Strong evidence of field-evolved resistance to the Bt toxins in transgenic crops was reported for some populations of three noctuid insects the stem borer, Busseola fusca, H. zea, and S. frugiperda (Matten et al. 2008). Field-evolved resistance of S. frugiperda to Bt corn producing Cry1F occurred in 4 years in Puerto Rico, USA (Matten et al. 2008). This was the first case of resistance leading to withdrawal of a Bt crop from the market (Matten et al. 2008). Field-evolved resistance of Bt corn producing Cry1Ab was found in a population of the stem borer, B. fusca, in South Africa in 8 years or less (Van Rensburg 2007). A second resistant populations of B. fusca to Bt corn was detected in another area in South Africa (Kruger et al. 2009). The percentage of farmers reporting medium or severe damage to Bt corn from B. fusca rose from 2.5% in the 2005–2006 growing season to 58.8% in the 2007–2008 season. In the USA, field-evolved resistance of H. zea to Bt cotton producing Cry1Ac was noticed in some populations of the insect in 7–8 years in the southeastern USA (Luttrel and Ali 2007). In China, evidence of field-evolved resistance to Cry1Ac expressing Bt cotton was detected in populations of H. armigera (Liu et al. 2010).

In contrast, strong evidence of sustained susceptibility to the Bt toxins in transgenic crops was reported for populations of 8 target insects on Bt corn and Bt cotton after 4–8 years. These insects were H. armigera, H. virescens, H. punctigera, P. gossypiella, D. grandiosella, D. saccharalis, O. nubilalis, and Sesamia nonagrioides (Tabashnik et al. 2009). However, In November 2009, Monsanto Company declared that P. gossypiella could develop resistance on Bt cotton producing Cry1Ac in four districts in India. As a solution for this problem the company produced another Bt cotton expressing Cry1Ac + Cry1Ab (Bagla 2010).

Safety of Bt crops

According to companies, like Monsanto, which produce genetically engineered crops containing Cry toxins, such toxins are supposed to be active only against particular insects and should have no deleterious effect on the environment or on mammals and humans (Mendelshon et al. 2003).

Safety to environment

Most of the Cry proteins deposited into soil by Bt crops were degraded in soil within a few days, and they had no effect on soil bacteria, actinomyces, fungi, protozoa, algae, nematodes, or earthworm. Bt corn or Bt cotton were found to have no significant effect on populations of beneficial insects. In addition, the remains or pollen of Bt crops had no hazards to the non-target plants in the fields of Bt crops (Mendelshon et al. 2003).

In laboratory studies at Cornell University (USA) in 1999, it was found that the pollen in Bt corn had deleterious effects on larvae of the Royal (or Monarch) Butterfly, Cithoroni aregalis (Losey et al. 1999). However, Proceedings of National Academy of Science revealed that the results of six laboratory and field studies showed that the density of Bt toxin in Bt corn pollen is not enough to cause any harm to the insect larvae (Sears et al. 2001). It is to be noted that monarch butterfly has a beautiful coloration with about 15-cm wing-span width and it is a matter of home-decoration in the USA.

Effect on the secondary pests

Lu (2010) reported that annual cultivation of Bt cotton resulted in high infestation levels by the sucking mirid insects in China which became the key pest on Bt cotton. Similarly, the continuous cultivation of Bt cotton caused obvious infestation by aphids and mealybugs in India (Losey et al. 1999). Laboratory tests conducted by Liu et al. (2005) showed that Aphid gossypii fed on Bt cotton had shorter reproductive duration, maximum lifespan, and an earlier peak of daily mortality in the 1st and 2nd generations compared to individuals fed on non-Bt cotton.

In addition, Lu et al. (2012) reported that after 20 years (1990–2010), a remarkable decline in aphid populations was noticed in Bt cotton fields in 36 locations in 6 districts north of China.

Safety to predacious insects

In laboratory studies, Mendelshon et al. (2003) found that pollen containing Cry toxins, which was at relatively very high doses, was not toxic to lady beetles (Coccinellids), green lacewings (Chrysoperla spp.), or honeybees. Also, field studies revealed that beneficial arthropods were substantially more abundant in Bt crops than in crops treated with chemical pesticides. Lu et al. (2012) reported a remarkable decline in aphid populations in Bt cotton fields in 36 locations in 6 districts north of China. They related this decline to the increase of the populations of the coccinellids, chrysopids, and spiders. In addition, these increased populations of the predators on Bt cotton had a considerable role for insect biological control on cotton, corn and peanut crops adjacent to Bt crops.

In another field study it was found that the populations of prevailing predators in a Bt corn field did not differ significantly from those on a conventional corn field. These predators were Hyppodamia convergens, Orius insidiosus, and Scymnus spp. (Al-Deeb and Wilde 2003).

A 6-year field study assessed the long-term impact of Bt cotton producing Cry1Ac toxin on 22 species and strains of foliar-dwelling natural enemies in Arizona (Naranjo 2005). The study revealed no chronic, long-term effects of Bt cotton on such natural enemies.

A 3-year field study was carried out by Moar et al. (2004) in the USA to estimate the effects of Bt cotton (Bollgard) on biological control agents. They concluded that there were no adverse effects on non-target arthropods (parasitoids and predators) in Bollgard cotton fields compared to conventionally grown cotton ones.

In Egypt, Dahi (2013) reported that Bt cotton producing Cry1Ac and Cry2Ab did not affect the populations or abundance of common predators species prevailing in cotton fields.

In 2009, Angelika Hilbeck’s team (ETH) in Zurich (Schmidt et al. 2009) published laboratory findings indicating that larvae of the two-spot ladybird, Adalia bipunctata, can be harmed by Bt toxins. The publication played a key role in justifying the cultivation ban for Bt maize MON810 in Germany imposed by Germany’s Environment Minister in April 2009 (Alvarez-Alfageme et al. 2010). In 2010, a paper was published by Jorg Romeis and his team at Switzerland’s Puplic Agroscope Researh Station in Zurich (Alvarez-Alfageme et al. 2010) which assessed the findings of Hilbeck’s group. The paper presented that the quantities of Bt toxins that ladybird larvae could be exposed to in the field are not expected to have any negative impact on such larvae.

In February 2012, the Hilbeck’s group published a further study (Hilbeck et al. 2012), in response to the 2010 publication of the Romeis group. They accused the Romeis group of using a different test method and this method was the reason for the difference in the results. They mentioned that combining the test methods from both groups showed that Bt toxin can indeed have a harmful effect on two-spot ladybird larvae. The result of Hilbeck’s group (Schmidt et al. 2009) was obtained by feeding ladybird larvae on eggs of the Mediterranean flour moth, Ephestia kuehniella that had been sprayed with Bt toxin solutions at different concentrations. They then found a higher mortality rate among the treated larvae compared to the control group. Romeis group examined ladybird larvae under the microscope and found that the larvae only suck the contents of the eggs and do not eat even a part of the egg shell which was sprayed with the Bt toxin (Alvarez-Alfageme et al. 2010). Consequently, the larvae are not exposed to the Bt toxin. In 2012 paper, Hilbeck’s group also observed ladybird larvae under the microscope and mentioned that the larvae bite into the eggs and when the contents spill out they come into contact with the egg shell sprayed with the toxin (Hilbeck et al. 2012). However, they added that in the field, ladybird larvae are only be exposed to potentially harmful quantities of Bt toxins if the feed on Bt corn pollen or on prey, except aphids, accumulate Bt toxins aphids only suck plant sap which does not contain Bt toxin.

In another study, the red spider mite, Tetranichus urtica was used as food for the larvae of A. bipunctata. The red spider mite is a natural prey of the ladybird and was fed on Bt corn before exposed to the larvae. The mortality rate of the treated larvae was not significantly different from that of the control group which was fed on conventional corn. This study confirmed that Bt crops are not harmful to the two-spotted ladybird (Romeis et al. 2012).

Hilbeck et al. (1998), in laboratory studies, fed 2nd and 3rd larval instars of the predator, Chrysoperla carnea on artificial diet mixed with Cry1Ab toxin. They found that the total mortality in larvae was significantly higher (57%) than in the untreated control (30%). Also, significantly more larvae died (29%) when received Cry1Ab later during their development compared to the control ones (17%). Although mortality was higher, almost no differences in developmental time were observed between treated and untreated larvae. In another study (Hilbeck et al. 1999), almost similar results were obtained when C. carnea larvae were fed on Spodoptera littoralis larvae fed on diet mixed with Cry1Ab and Cry2A at different concentrations.

However, Moussa et al. (2018) reported that feeding larvae of Chrysoperla carnea on aphids reared on Bt corn until pupation did not affect percentages of pupation or adult emergence of the predator.

Safety to honeybees

Laboratory feeding studies carried out by Rose et al. (2007) showed no effects on the weight and survival of honeybees fed on Cry1Ab sweet corn pollen for 35 days. In field studies, colonies foraging in sweet corn plots and fed on Bt pollen cakes for 28 days showed no adverse effects on bee-weight, foraging activity and colony performance. Brood development was not affected by exposure to Bt pollen. Feeding the 2nd instar larvae on pure Bt toxins mixed with their food on concentrations far above those to which they would be exposed showed insignificant mortality rate between the treated larvae and the control group.

Duan et al. (2008) examined 25 studies that independently assessed potential effects of Bt toxins on honey bee survival and found that the Cry proteins did not negatively affect the survival of either honey bee adults or larvae in laboratory.

Are Bt crops safe to mammals and humans?

World controversy

The results of experiments conducted by researchers at the University of Caen, France, and supported by GEKKO Foundation, in Germany, showed that toxins produced in Bt corn, Mon810, can impact significantly the viability of human cells. The effects were observed with relatively high concentrations of the toxins however, further investigations should be conducted to find out how such toxins impact the cells. In addition, it should be taken into account if there are combination effects with other compounds in the food (Mesnage et al. 2011). By introducing the toxin-gene into the plant, the structure of the toxin is modified and may cause its selectivity to be changed. Many Bt corn, like Smart Stax, produce 2–6 different Bt toxins and therefore have a higher content of toxins (Mesnage et al., 2012).

In 2001, the Environmental Protection Agency (EPA) in the USA supervised comprehensive studies to reassess the 4 registered Bt crops that had been accepted for agricultural use since 1995. These crops were Bt corn (Cry1Ab), Bt corn (Cry1F), Bt cotton (Cry1Ac), and Bt potato (Cry3A). The reassessment included the potential effects on the environment, the natural enemies, the non-target insects and the safety to human and mammals (Mendelshon et al. 2003). The results of the reassessment indicated that in vitro studies Bt toxins were unstable in the presence of digestive fluids of human’s gut and were degraded in such fluids within 0–7 min. However, these studies did not ensure the non-toxicity of these toxins to human or that the rapid degradation occurs in all Cry toxins. The Cry1Ab and Cry1Acin processed corn foods (popcorn, tachoshell, cornflex, starch, oil, etc.) are not heat-stable and accordingly become inactive in such foods. No acute toxicity was shown in mice treated with high doses of Bt toxins, 3280–5000 mg/kg body weight.

Wang et al. (2002) found that feeding mice on Bt rice flour (Cry1Ac) at a dose of 64 mg of the toxin/kg body weight for 90 days did not cause any effect in the tissues of the liver, kidney, intestines, or blood cells. In addition, no significant differences in the weights of such organs between treated and untreated mice.

Hall (2011) mentioned that the risks of Bt foods to human health appear small based on what is known about the bacterial endotoxin, its specificity and confidence of the process of plant transformation and screening. The tasks of determining the levels of such risks, however, are immense. Human diets are complex and variable, so, how can we trace the acute or chronic effects of eating Bt foods when they are mixed with many other foods that may also present their own health hazards? It is even more complicated to determine the indirect risk of eating meat from animals raised on transgenic crops. These tests take time and the results of clinical trials are not always clear-cut. It will likely take decades before knowing with any certainty if Bt crop is safe for human or not.

An analysis of blood and organ system was carried out with rats fed three main commercialized Bt maize (deVendomois et al. 2009). Approximately 60 different biochemical parameters were classified per organ and measured in serum and urine after 5 and 14 weeks of feeding. Bt maize-fed rats were compared first to their respective parent non-Bt equivalent control groups, followed by comparison to 6 groups which had consumed other non-Bt maize varieties. The analysis clearly revealed sex- and dose-dependent effects on the kidney and liver of the treated rats. Other effects were also noticed in the heart, adrenal glands, spleen, and hematopoietic system.

The French High Council of Biotechnologies Scientific Committee reviewed the 2009 Vendomois et al. study and concluded that “It presents no admissible scientific element likely to ascribe any hematological, hepatic, or renal toxicity to the three re-analyzed Bt maize (Anonymous 2010).” Also, a review by Food Standards Australia New Zealand of the 2009 Vendomois et al. study concluded that the results were due to chance alone. However, French government applied a principle of precaution against genetically modified crops. In addition, a review by Food Standards Australia New Zealand of the 2009 Vendomois et al. study concluded that the results were due to chance alone (Anonymous 2010).

A Canadian study in 2011 estimated the presence of Cry1Ab1 (Bt toxin) in non-pregnant women, pregnant women, and fetal blood. All groups had detectable levels of the toxin in blood, including 93% of pregnant women and 80% of fetuses at concentrations of 0.19 ± 0.30 and 0.04 ± 0.04 ng/ml, respectively (Anonymous 2010).

In 2004, a human feeding study was conducted to determine the effects of genetically modified (GM) food. Seven human volunteers were allowed to eat genetically modified soybean (resistant to the herbicide Roundup) to see if the DNA of GM soybean was transferred to the human gut bacteria. The examination of their guts showed that no recombinant DNA was found (Netherwood et al. 2004). However, the anti-GM crops advocates believe that the study needs additional testing to determine its significance (Smith 2007).

In a study funded by the European Arm of Greenpeace, it was found that there was a possibility of a slight but statistically meaningful risk of liver damage in rats (Seralini et al. 2007). However, this possibility of risk was reported to be of no biological significance by the European Food Safety Authority (Seralini et al. 2007).

Anilkumar et al. (2010) fed sheep (1 year old) on Bt and non-Bt cottonplants for 3 months and found that the histological examination of liver and kidney revealed no significant changes between Bt and non-Bt plant-fed sheep.

In a study in pigs and calves fed on Bt maize, it was found that Cry protein fragments were detectable, but reduced in size, as they travel down the gastrointestinal (GI) tract. None were detected in the liver, spleen, or lymph nodes indicating that they were too large to be systematically absorbed from the GI tract (Chowdhury et al. 2003). It was suggested that transgenic nucleic acid and proteins from GM crops are handled in the gut like their conventional counterparts, with no evidence for systemic absorption of intact proteins or genes (Sieradzki et al. 2013). Herman et al. (2006) reported that maize fed to animals is generally not processed and accounts for approximately 65% of their diet. As for human, the exposure to Cry protein is much lower than that of farm animals and the maize is processed by heating causing the Cry proteins to lose their insecticidal activity and make them more susceptible to degradation.

A number of Cry proteins (toxins) were subjected to in vitro heat stability studies under conditions similar to those used for human food processing (Hammond and Jes 2011). All Cry proteins tested lost insecticidal activity after processing. In general, there are fundamental biological properties of proteins that greatly limit their potential to produce chronic toxic effects when ingested (Hammond et al. 2013). The ingestion of proteins introduced to date into GM crops are not considered to be toxic based on their known biochemical function and on the results obtained from bioinformatics searches (Hammond et al. 2013).

In a safety study (Onose et al. 2008), rats were treated with famotidine (to reduce gastric acid secretion) and indomethacin (to damage the intestinal epithelium), then they were fed diets with and without Cry1Ab protein (10 ppm). Despite the expectation of less Cry1Ab protein digestion and more absorption of Cry1Ab protein into the circulatory system of the GI-impaired rats, there was no evidence of toxicological effects (changes in clinical blood parameters and histologic appearance of organs) in the treated rats.

Mezzomo et al. (2013) reported that Bt spore preparations containing various Cry proteins were found to cause hemato-toxocity in mice when administered by oral gavage. However, Koch et al. (2015) mentioned that such a result could be due to spore components other than Cry proteins.

Domingo (2016) stated that with only a few exceptions, the reported studies in the last 6 years showed rather similar conclusions the assessed GM soybeans, corn, rice, and wheat would be as safe as the parental species of these plants. However, in spite of the notable increase in the available information, studies on the long-term health effects of GM plants, including tests of mutagenicity, teratogenicity, and carcinogenicity seem to be still clearly necessary.

StarLink corn and human allergy

StarLink is a variety of Bt corn produced commercially for use in animal feed by Aventis Company in the USA. There was a stipulation that the crop must not be used for human consumption because the Bt toxin used in StarLink is less rapidly digested than the other Bt toxins. Twenty-eight people showed allergic reactions related to eating corn products that may have contained the StarLink toxin. However, the US Centers for Disease Control studied the blood of these people and concluded that there was no evidence that the allergic reaction was related to the StarLink toxin (Seralini et al. 2007).

Ban of Bt crops

In August 2003, Zambia cut-off the flow of genetic modified maize from UN’s World Food Program. This left a famine-stricken population without food aid. In December 2005, the Zambian government changed its mind in the face of further famine and allowed the importation of GM maize (Anonymous b, 2009).

In April 2004, Hugo Chavez announced a total ban on GM seeds in Venezuela (Anonymous, 2009).

In January 2005, the Hungarian government announced a ban on importing and planting of GM maize seeds, which was subsequently authorized by the European Union (James 2016).

The Germany’s Environment Minister issued a ban for cultivation of Bt maize in April, 2009, after published laboratory findings in Germany indicating that A. bipunctata larvae can be harmed by Bt toxins (Schmidt et al. 2009).

In 2014, the Minister of Agriculture in Egypt announced a ban on the cultivation of Bt cotton and maize after a debate on a TV program concerning the possibilities of hazards of Bt crops (unpublished).

In Hawaii, growing Bt cotton has been prohibited since 2013 (James 2016).

Burkina Faso, Africa top cotton producer, banned Bt cotton in 2016 because of economic and quality concerns (James 2016).

Romania decided not to plant GM crops in 2016 due to onerous requirement by the government (James 2016).

In 2015, European Commission announced that 19 EU countries are able to ban the cultivation of GM crops. Although repeated scientific assessments have concluded that GM crops are as safe for humans and environment as the conventional crops, a majority of Governments, parliamentarians, and European people oppose cultivation of such GM crops (James 2016). The European countries that banned cultivation of GM crops made their decision because they prefer producing the organic food. These countries import over 30 million tons per year of Bt corn and GM soybeans as animal feed and livestock industry. However, Russia issued a ban on both cultivation and importing Bt crops (James 2016).

Monsanto declared worst company of 2011

Monsanto, a major biotech corporation responsible for genetically modified foods, has been given “the Worst Company of 2011 Award” for threatening both human health and the environment. The award was given by natural health information website Natural Society after thousands of readers voted “Monsanto the worst company of 2011” (Gucciardi 2011). Numerous scientific studies have found Monsanto’s GM crops, herbicides and bio-pesticides, to be a danger to the planet. A review of 19 studies announced that consumption of GM corn or soybeans lead to significant organ disruptions in rats and mice, particularly in the liver and kidney (Gucciardi 2011).

The end of Bt crops

A report from Texas A & MA griLife Extension stated that Bt cotton and corn have been attacked by bollworms and earworms among other pests. Cry toxins had a good run and will hang on for a while longer, but the era of the Cry toxins seems to be ending. They suggested that Bt crops should contain two or three different toxins to delay resistance. If an insect had an allele to survive on toxin 1, it probably does not have different alleles to survive on toxins 2 and 3 (Anonymous 2016).

The report also claimed that the newest Vip (Vegetative insecticidal protein toxins) for caterpillars does a good job of controlling many species. Once again, the insects will have adapted, or partially adapted, to the old toxins, so selection for resistance will be on Vip and again the era of the Cry toxins seems to be ending (Anonymous 2016).

No to a moratorium on the cultivation of GM maize

A declaration of 500 European scientists indicated that amoratorium on the cultivation of GM Bt maize approved by the European Union (EU) is not scientifically justified. Such a decision could be based only on imaginary or false uncertainties concerning environmental or food safety. It would bring no new knowledge that could reduce the hypothetical risks that could be generated by the cultivation and the consumption of GM Bt maize. Such a moratorium would be in contradiction with the precautionary principle (Naud et al. 2007).

Plant varieties obtained either by classical breeding or by gene transfer share a similar level of risk, which is low in both cases. Maize has been cultivated and eaten by humans for thousands of years without any negative effect for animal or human health despite the numerous genetic modifications undertaken during classical genetic selection. The insecticidal active toxin presents in Bt maize has been exploited for decades within the applied commercial compounds of Bacillus thuringiensis without any observed toxicity or allergic response (Naud et al. 2007).

Bacteria from natural populations transfer plasmids mostly towards their kin

Plasmids play a key role in microbial ecology and evolution, yet the determinants of plasmid transfer rates are poorly understood. Particularly, interactions between donor hosts and potential recipients are understudied. Here, we investigate the importance of genetic similarity between naturally co-occurring Escherichia coli isolates in plasmid transfer. We uncover extensive variability, spanning over five orders of magnitude, in the ability of isolates to donate and receive two different plasmids, R1 and RP4. Overall, transfer is strongly biased towards clone-mates, but not correlated to genetic distance when donors and recipients are not clone-mates. Transfer is limited by the presence of a functional restriction-modification system in recipients, suggesting sharing of strain-specific defence systems contributes to bias towards kin. Such restriction of transfer to kin sets the stage for longer-term coevolutionary interactions leading to mutualism between plasmids and bacterial hosts in natural communities.

1. Introduction

Conjugative plasmids play a central role in horizontal gene transfer, impacting both evolutionary and ecological processes. At large phylogenetic scales, they are the main vector of genetic exchange among bacteria [1], shaping gene flow and long-term adaptation of communities. They also encode a diversity of ‘accessory genes' [2] often conferring environment-specific adaptations such as antibiotic and metal resistance and virulence traits. As a consequence, the dynamics of horizontal transfer has crucial consequences for the outcome of competition between lineages, which in turn can both drive the epidemiology of bacterial pathogens and influence ecosystem services. In particular, antibiotic-resistance-conferring plasmid transfer can govern success of strains within patients [3–5] and facilitate pathogen epidemics [6]. An understanding of factors controlling transfer rates is therefore critical.

A striking feature of plasmid transmission by conjugative transfer is its variability. Indeed, estimates of transfer rates lead to fundamentally different conclusions about whether plasmids can naturally persist in the absence of selection on plasmid-carried traits [7–10]. Transfer rates are dependent on both the initiation of conjugation in donor cells and successful establishment in recipient cells [11]. Quantification of transfer rates in the laboratory has mostly focused on a few laboratory strains, which are poor models for natural populations [12], despite the strong effect host genotype and plasmid–host interactions can have on plasmid transfer rates. Transfer rates for plasmid R1 span seven orders of magnitude among natural isolates [13,14]. Plasmids might thus be either lost or spread to fixation depending on host community composition—with rare efficient donors having a particularly strong effect [14]. The detailed pathways of plasmid transfer can also have profound implications. Different plasmid groups are specifically associated with different host lineages, suggesting that barriers to transfer can contribute to global patterns in plasmid host range [15–18]. Moreover, biased transfer of beneficial plasmids towards kin (i.e. between donors and recipients of the same genotype) can favour host bacteria with high investment in transfer, through kin selection [19]. This in turn could lead to higher transfer for traits including antibiotic resistance under antibiotic selection.

Earlier work showed that bacterial hosts from strain collections of Escherichia coli indeed display biased transfer to kin [19]. However, isolates from these collections were distantly related, and unlikely to have coexisted in natural environments. To understand its ecological and evolutionary implications, the diversity in transfer rates among bacteria needs to be characterized within natural populations. Genetic variation in transfer rates, or genetic distance effects, might exist only at a large phylogenetic scale or be the product of environment-specific selective forces, such that strains isolated from the same environment have homogeneous transfer rates or display no bias in transfer towards kin. Here, we investigate the transfer rates of two resistance plasmids, R1 and RP4, with, respectively, narrow and broad host ranges, among a collection of E. coli natural isolates, for which population structure and native plasmid content have been characterized previously [16]. Escherichia coli from independent populations of grazing cattle were shown to display hierarchical population structure, with high variability in genotypes across cattle populations and individuals, but reduced variability within individuals [16]. Plasmid carriage varied within serotype within individual cattle, suggesting that most opportunity for plasmid transfer was between closely related bacteria. We ask if the striking diversity in transfer previously observed within heterogeneous laboratory strain collections is still present within natural populations, when host bacteria are isolated from the same natural population or are closely related, and we explore genetic factors that could contribute to biased transfer towards kin.

2. Methods

(a) Bacterial strains and plasmids

Escherichia coli field isolates (electronic supplementary material, table S1) were selected from the field collection characterized in [16], which surveyed E. coli diversity and plasmid content in grazing cattle. Genotypic diversity was previously assessed by sequencing H-antigens classifying E. coli into serotypes [20]. Serotypes were highly diverse across individual cattle and cattle populations by contrast, within-host diversity was reduced, with most cowpats containing only one or two serotypes [16]. To cover E. coli diversity present in the collection, we selected 14 strains belonging to seven different serotypes and originating from six different field sites (electronic supplementary material, table S1). Isolates with detected plasmid replicons were excluded. We also used laboratory strains of the E. coli K-12 lineage, MG1655 and its derivative MFDpir [21], as standard recipient and donor strains. To test for the effect on transfer rates of restriction-modification (RM) systems, defence systems against foreign DNA [22], we used MG1655 ΔhsdS::Kn generated by P1 transduction from the Keio collection [23].

We characterized transfer rates for two plasmids belonging to the replicon incompatibility groups most abundant in the field collection [16]: IncF, a group of narrow-host-range plasmids only replicating in Enterobacteriaceae, and IncP, a group of broad-host-range plasmids that can transfer and replicate in a wide range of Gram-negative bacteria. R1 plasmid [24] is an IncFII plasmid, in which regulation of transfer is representative of the majority of IncF plasmids [25]. RP4 plasmid [26] is a model IncP-α plasmid. R1 and RP4 were conjugated into unmarked field isolates using the donor strain MFDpir [21], which requires di-aminopimelate (DAP, Sigma-Aldrich) to grow in LB medium. For all other conjugation assays, spontaneous rifampicin-resistant (Rif R ) mutants of MG1655 and the 14 field isolates were generated by plating overnight cultures on LB-agar with rifampicin (Rif, Sigma-Aldrich) at 100 µg ml –1 , to use as recipients.

(b) Experimental design and conjugation assays

Conjugation assays were performed by mixing equal volumes of overnight cultures of donors and recipients with 10-fold total dilution into 1 ml LB medium, supplemented with DAP 0.3 mM when MFDpir was used as a donor. Overnight cultures for conjugation experiments did not contain antibiotics. Mixes were incubated at 37°C with 150 r.p.m. shaking. To favour detection of relatively low transfer rates, mating assays were performed for 3 h as a general standard. When stated, assays with a reduced 1 h mating were performed to limit secondary transfer from transconjugants. Donor, recipient and transconjugant densities were then estimated by dilution plating onto selective plates: plasmid-containing bacteria were selected with kanamycin (Kn, 50 µg ml −1 ), except in experiments including MG1655 ΔhsdS::Kn strain and R1 plasmid, for which chloramphenicol (Cm, Sigma-Aldrich, 25 µg ml −1 ) was used instead (see electronic supplementary material, figure S1 for details). Control assays (with donors only or recipients only) never showed any growth on selective transconjugant plates. Each conjugation was performed with at least two biological replicates (separate conjugation assays) per experiment and two independent experiments done on different days.

We first performed conjugation assays from K-12 to the 14 unmarked field isolates, to estimate standard recipient ability and generate plasmid-bearing field isolates. Assays from plasmid-bearing isolates to the Rif R K-12 recipient then estimated standard donor ability. Next, we used pairs of isolates as donor and recipients (electronic supplementary material, table S1). We first compared 14 kin pairs (where the recipient is the Rif R derivative of the donor) and 14 non-kin pairs where donor and recipient differ in both serotype and isolation site. For R1 plasmid, we performed additional assays with 14 pairs sharing serotype but isolated from different sites and 14 pairs from different serotypes isolated from the same site. Finally, to test if RM systems contribute to transfer patterns we compared transfer rates of our field isolates towards the standard K-12 recipient (RM + ) and its mutant with no functional type I RM system (RM − ), MG1655 ΔhsdS::Kn [27]. RM systems combine a restriction enzyme that cleaves a specific DNA sequence, and a cognate methyl-transferase protecting that same sequence. Foreign DNA originating from cells lacking an RM system present in recipients is not methylated, and thus targeted by restriction. If restriction based on K-12 RM system limits transfer from our natural isolates, we expect transfer rates towards the RM − mutant to be higher than towards the RM + recipient.

(c) Isolate genotyping and phylogenetic distance

We used the phylogenetic markers identified in [28] to study phylogenetic relationships among isolates. The three markers dinG, DPP and tonB were amplified from the 14 field isolates with PCRBio Taq Mix Red (PCR Biosystems) using primers described in [28], and sequenced through Eurofins Genomics. For K-12 strains MG1655 and MFDpir, sequences from GenBank accession NC_000913 were used. Sequence data were pre-processed in G eneious (v. 8.1.6). Amplicons were trimmed at both 3' and 5′ ends to remove low-quality sequences (i.e. base pairs with an error probability above 5%). High-quality alignments (respectively, 854, 792 and 725 bp long for dinG, DPP and tonB) were concatenated and used to determine multi-locus phylogenetic distance. For isolates D7.8 and oc5.1, tonB primers did not yield any product DPP and dinG products revealed both isolates were part of E. marmotae species. Phylogenetic trees were made with a weighted neighbour-joining tree building algorithm implemented in G eneious . To obtain phylogenetic distances among all isolates including E. marmotae, the tree built with DPP and dinG products only was used distances obtained were highly correlated with the ones using all three gene products within E. coli isolates (Spearman correlation coefficient ρ = 0.98, p < 2.2 × 10 −16 ). Spontaneous Rif R mutants were considered identical to the strain they originated from (genetic distance of 0).

(d) Data analysis

Conjugation rates were measured as γ = T / D R t (ml cell −1 h −1 ), where T, D and R, respectively, indicate the density of transconjugants, donors and recipients (cells ml −1 ), and t indicates incubation time (h). When no transconjugants were detected, a threshold conjugation rate was calculated by assuming that one single transconjugant colony was observed. For assays using field isolates as both donors and recipients, variable growth was observed, particularly for recipients (electronic supplementary material, figure S2), probably owing to the spontaneous Rif R recipients used in these assays. In order to limit variation in computed conjugation rates owing to low donor or recipient densities, data were excluded when either recipient or donor densities were less than 2 × 10 7 cells ml −1 . As transfer rate values spanned several orders of magnitude, all statistical analysis used log10-transformed data, and averages across replicates were computed as geometric means. For standard donor and recipient ability assays, the effect of field strain identity on transfer rates was tested with one-way ANOVAs. The effect of relationship between donor and recipient within the strain collection was tested with type I ANOVAs as transfer rate∼donor identity + recipient identity + relationship. When testing for RM effects, the effect of recipient strain RM status was tested with a type I ANOVA as transfer rate∼donor identity × recipient identity. R v. 3.4.1 was used for all analyses [29].

3. Results

(a) Variation in donor ability and recipient ability across natural isolates

To analyse the amplitude of variation in transfer rates, we first measured transfer rates using K-12 laboratory strains as standard donor or recipient. One of 14 isolates, D2.2 was observed to repeatedly kill K-12 strains in each assay (with less than 1% of K-12 inoculum detected after mating), it was thus excluded from analysis.

For both R1 and RP4 plasmids, transfer rates spanned more than five orders of magnitude overall, from around 10 −16 –10 −15 ml cell −1 h −1 (detection threshold) to more than 10 −10 ml cell −1 h −1 (figure 1). For the isolates tm8.6 and R1.9, we were unable to obtain any clones with RP4 across several assays, revealing very low recipient ability donor ability was not quantified. For both plasmids, donor and recipient ability of K-12 was always as high or higher than any of the field isolates. Excluding K-12, recipient genotype significantly affected conjugation rate from a standard donor for both R1 (F12,39 = 16.05, p < 2.10 −11 ) and RP4 (F12,39 = 13.1, p < 10 −9 ). Similarly, excluding K-12 donor genotype significantly affected conjugation rate towards the standard recipient for both R1 (F12,37 = 11.3, p < 10 −8 ) and RP4 (F10,33 = 8.07, p = 2.10 −6 ). The lowest amplitude of variation was observed for R1 plasmid donor ability, for which all measured rates were above 10 −14 ml cell −1 h −1 . We hypothesized that efficient secondary transfer from K-12 recipients was masking actual variability in transfer from primary donors. Measuring conjugation rates with reduced mating time revealed higher variation among isolates (electronic supplementary material, figure S3) and a stronger effect of the donor (F12,37 = 31.8, p < 10 −15 ) than with longer mating.

Figure 1. Extensive variation in plasmid donor and recipient ability across field isolates of Escherichia coli. Conjugation rates were measured in liquid with shaking over 3 h, towards K-12 MG1655 Rif R (donor ability) and from K-12 MFDpir (recipient ability). Individual replicates are shown as dots, lines are geometric means. Rates with K-12 as both donor and recipient are shown on the left field isolates are then ordered by their phylogenetic distance to K-12. Colour indicates site of origin for each isolate. A tree showing phylogenetic relationships is shown under strain names.

No correlation was observed among isolates between average donor and recipient ability (Spearman rank-correlation ρ = 0.02, p = 0.92) or between average rates between plasmids R1 and RP4 (Spearman rank-correlation ρ = 0.35, p = 0.08). Moreover, phylogenetic distance from the standard K-12 donor or recipient did not explain average conjugation rates towards or from natural isolates (electronic supplementary material, figure S4 transfer rate∼genetic distance R 2 = 0.003, F1,48 = 0.14, p = 0.71).

(b) Comparing transfer among kin and non-kin

We next analysed diversity in transfer rates within our natural populations. To test if transfer is increased towards kin (defined as recipients with strict genetic identity to donors), we performed for each donor conjugation assays to a marked recipient of the same isolate (transfer to kin) and to a randomly chosen isolate both belonging to a different serotype and isolated from a different field site (transfer to non-kin). Rates of transfer were strongly variable across isolates, spanning five orders of magnitude from less than 10 −6 to greater than 10 −12 ml cell −1 h −1 (figure 2 electronic supplementary material, figure S5). The identity of donor and recipient had a strong effect on conjugation rates (for R1 plasmid, donor effect F13,70 = 20, p < 2 × 10 −16 , recipient effect F9,70 = 10.9, p < 3 × 10 −10 for RP4 plasmid, donor effect F11,74 = 5.93, p < 10 −6 , recipient effect F7,74 = 11.5, p < 10 −9 ). However, even after accounting for both donor and recipient identity, the relationship between isolates (i.e. being kin or non-kin) was the factor with the largest effect on conjugation rates (for R1, F1,70 = 108, p < 10 −15 for RP4, F1, 74 = 34.1, p < 2.10 −7 ). On average, a given donor transferred plasmid R1 towards kin 29-fold more efficiently than towards non-kin, and plasmid RP4 40-fold more efficiently. This effect was highly variable across couples, but no isolate was observed to transfer at significantly higher rates towards non-kin. Moreover, higher transfer towards kin could not be explained by an effect of kin on cell densities during competition, as cell densities were not significantly different when donor and recipients were kin or non-kin (one-way ANOVA, donor density∼relationship, F1,186=0.018, p = 0.89, and recipient density∼relationship, F1,186=0.32, p = 0.57). There was still high variability among isolates considering only transfer towards clone-mates (transfer rate∼strain identity, for R1 F13,35 = 14.1, p = 3.10 −10 , for RP4 F11,33 = 9.09, p < 10 −6 ), spanning five orders of magnitude. When the same couples of isolates were tested for both plasmids, no correlation in average transfer rates between R1 and RP4 was observed across couples (Pearson correlation coefficient r19 = 0.17, p = 0.46).

Figure 2. Variable transfer rates among field isolates and bias towards kin. Mating assays were performed for 3 h, with donors containing either R1 or RP4 plasmid. Pairs are shown ordered by donor isolate, with the recipient isolate being the same as the donor (kin, red) or another field isolate with distinct serotype and isolation site (non-kin, light blue, see electronic supplementary material, table S1 for recipient identity). Individual replicates are shown as dots, lines are geometric means. Summary graphs at the right show average transfer per couple of strains (dots) and overall geometric means per treatment (lines). Electronic supplementary material, figure S5 presents the same data ordered by recipient isolate.

(c) Effect of genetic distance and field site on transfer rates

To understand what leads to the higher conjugation rates observed among clone-mates, we tested if high transfer rates required strict kin identity (genetic distance = 0), or if transfer rates gradually increased with genetic proximity. We focused on the R1 plasmid representative of the IncF plasmid group most abundant in the strain collection, and performed additional conjugation assays choosing couples of isolates that share serotype (initial assessment of their relatedness) or field site of isolation (electronic supplementary material, table S1). We asked if similar variation is observed with shared serotype or within sites, or if isolates from the same field or with same serotype transfer preferentially to each other (figure 3a). Overall, the relationship between donors and recipients (i.e. clone-mates, same serotype, same field or no relation) still affected transfer (relationship effect F3,142 = 44.95, p < 2 × 10 −16 ). However, post hoc Tukey tests showed that the only relationship that was significantly different from others was clone-mates, with higher transfer rates (p < 10 −7 for all comparisons to clone-mates, p > 0.5 for all others). Isolates from the same field sites also showed variation in transfer rates (e.g. site R, electronic supplementary material, figure S6), confirming that diversity in transfer rates does occur within the natural populations sampled.

Figure 3. Higher transfer rates among field isolates are not correlated with phylogenetic distance but strictly restricted to kin (clone-mates). The average R1 plasmid conjugation rate for a given couple of donor and recipient isolates is shown as a function of the relationship initially defined between donor and recipient (a), and of the genetic distance between donor and recipients (c). Average rates per couple are shown as dots boxplots in (a) show average rate and 95% confidence interval for each relationship. (b) Phylogenetic relationships between 14 field isolates and K-12 laboratory strain, with serotype numbers indicated in brackets.

The absence of a serotype effect implies that sharing serotype does not confer high enough relatedness to be equivalent to kin. To understand how genetic distance affects transfer rates more precisely, we derived phylogenetic distance among isolates, which revealed that serotype was a poor indicator of phylogenetic distance (figure 3b). Two isolates, D7.8 and oc5.1, were even identified as belonging to E. marmotae, despite sharing serotypes with E. coli isolates. Overall, there was a small but significant negative effect of phylogenetic distance on conjugation rates (transfer rate∼distance, estimate = −9.03 ± 2.86, r 2 = 0.05, p < 0.002). However, after considering kin/non-kin status, there was no additional effect of phylogenetic distance (transfer rate∼kin status + distance, distance estimate 0.65 ± 3.14, p = 0.836). Transfer rates were thus not linked to general phylogenetic similarity, but depended only on whether interacting couples were clone-mates or not. Among non-kin, both closely related and inter-species couples had transfer rates spanning from less than 10 −16 to greater than 10 −13 ml cell −1 h −1 (figure 3c), suggesting barriers to transfer can be present even among closely related genotypes, and clone-mates specifically have less barriers to transfer.

(d) Variation in restriction-modification systems as a mechanism for biased transfer

The restriction of high transfer rates to clone-mates suggests that barriers to transfer are caused by one or few genetic determinants variable at short phylogenetic scales. We tested if variation in RM systems can contribute to the barrier to conjugative transfer in these field isolates, by comparing transfer rates of field isolates towards the standard K-12 recipient (RM + ) and an RM − mutant. We first confirmed that R1 transfer within K-12 is affected by restriction (figure 4, left): as expected, the RM + strain transfers equally well to both RM + and RM − recipients the RM − strains transfers at the same rate towards itself, but transfer from the RM − strain is restricted in RM + recipients. When measuring transfer from field isolates (figure 4, right), in addition to a strong effect of donor isolate (donor effect F12,106 = 47.2, p < 2 × 10 −16 ), recipient RM status was also significant (F1,106 = 30.6, p < 3 × 10 −7 ). On average, the RM − recipient received R1 plasmid at 3.15-fold higher rates than the RM + recipient. However, the donor/recipient interaction was significant as well (F12,106 = 2.75, p = 0.003), with only some isolates transferring R1 more efficiently towards the RM − strain, as expected if some donors also bear an RM system with same specificity as K-12 type I RM. Our results indicate that R1 plasmid is efficiently targeted by restriction, and suggest variation in RM content among field isolates.

Figure 4. The K-12 type I RM system limits transfer from natural isolates. Mating assays were performed for 1 h, from R1 plasmid donors shown on the x-axis towards a K-12 recipient with (RM + , blue) or without (RM − , red) its native RM system. Individual replicates are shown as dots, lines are geometric means. Positive controls with K-12 donors are shown left of the dashed line: deactivating RM in donors decreases conjugation rate when recipients are RM-positive.

4. Discussion

We show here that variation in plasmid transfer within E. coli isolates from common environments is similar to the variation seen in strain collections [13,14], implying that such variation does not arise from different environment-dependent selective pressures. On the contrary, large differences in transfer rate persist within field site or for closely related isolates. Donor and recipient abilities, as well as transfer rates for R1 and RP4 plasmids, were not correlated, consistent with the different mechanistic basis and regulation of transfer operons in their respective plasmid classes [30]. Interestingly, the broad-host-range plasmid RP4 had similar variation in transfer among hosts, and was no less sensitive to host control than R1, despite suggestions that IncF narrow host range might arise from their more complex regulation by host cells [31]. Moreover, variation in transfer rates among natural isolates might even be higher than estimated here, as we selected isolates with no detected replicons, limiting the effect of modulation of transfer rates by co-resident plasmids [32].

In addition to donor and recipient identity, the main factor controlling transfer rates was the relationship between donors and recipients, with transfer towards kin (clone-mates) being more than 10-fold higher than towards non-kin. We therefore extend the pattern identified previously [19] to a second plasmid, the broad-host-range RP4. The average bias towards kin was even higher for RP4, consistent with the fact that it lacks anti-restriction genes present on R1 [33]. Importantly, we show that bias towards kin occurs among lineages coexisting in the field, indicating that this phenomenon is prevalent in natural populations. Moreover, the effect is restricted to close kin, with no higher transfer towards isolates with relatively closer genetic distance. Thus, discrimination towards kin is here not a function of average genetic distance among strains [34], but might arise from a combination of few loci [35], likely to be variable even at short genetic distances. Our results are consistent with a study on Dickeya strains isolated from the same field site, that despite not being genetically distinguishable using genomic fingerprints, displayed high variation in recipient ability [36].

We identified restriction-modification as a likely mechanism contributing to discrimination in transfer. Restriction was previously shown to limit plasmid conjugation rates with relatively low efficiency [22,37], likely because the first transconjugants escaping restriction are then protected from further restriction when transferring to kin. Similarly, the increase in transfer we observe when inactivating K-12 type I RM is significant but relatively weak in comparison to the strong effect of donor strain. RM systems have tremendously variable target sequence specificity [38], and expression of several systems has a multiplicative effect on restriction efficiency [39], which could amplify the effect we measure with a single system. Our results agree with studies describing the role of RM systems in restricting transfer among lineages [40,41]. As RM systems are very often mobile [42], their transfer among distant strains and loss among closely related strains could explain the large variation in transfer rates independent of genetic distance we observe. The other well-studied defence system of bacteria, CRISPR-Cas, appears less likely to explain our results: targeting of plasmid sequences by recipients could explain some genetic variation in recipient ability [43] but not why transfer is more efficient when plasmids are donated by kin. Some recently discovered mechanisms, however, BREX [44] and DISARM [45], have an epigenetic ‘memory' similar to RM systems, which might also contribute to preferential transfer to kin. Finally, other discrimination or structuring processes, not directly targeting plasmid conjugation, would also lead to discrimination in transfer if they affect how much donors encounter kin versus non-kin. This includes non-kin killing by bacteriocins, a form of kin discrimination [46]. Spatial structure, which promotes transfer to kin in the absence of discrimination mechanisms [19], can also bias transfer across a population. Indeed the E. coli populations sampled for this study show strong population structure, indicating that opportunities for transfer to plasmid-free isolates occur predominantly within genotypes [16].

The diversity in transfer rates that we uncover has consequences for understanding plasmid maintenance and ecological dynamics. The rates of transfer to kin vary here by five orders of magnitude. These transfer rates within lineages are one of the key determinants of plasmid maintenance [47]. Nine different plasmids were recently shown to be transferred at rates sufficient for persistence, in a classical K-12 strain [10]. Our results suggest that these conclusions should be taken with caution, as natural E. coli will probably transfer less than K-12. The scale of variation we observe implies that maintenance of plasmids might depend on subtle details of host genetic composition. Still, a few efficient donors can promote transfer in mixed bacterial populations [14], helping maintaining plasmids in mixed communities [48]. On the other hand, the biased transfer to kin we observe will limit that dynamic, and promote plasmid clustering in distinct lineages. This probably contributes to the variability in plasmid carriage observed among genotypes in the strain collection our strains originate from [16], and in pathogenic lineages [49]: high transfer rates from efficient donors will be mostly restricted to their own lineage, while strains with low transfer rate might not maintain plasmids efficiently, leading to ‘plasmid-shy’ genotypes [16]. Moreover, when plasmids confer benefits to their hosts, as with antibiotic exposure for antibiotic resistance plasmids, restricting transfer towards kin will benefit host bacteria and promote indirect selection of efficient donor hosts, through kin selection mechanisms [19]. Transfer towards non-kin, which is efficient for some pairs of isolates in our field collection, might also benefit the hosts when the transferred plasmids bear public-good-encoding genes [50,51], for instance virulence, antibiotic resistance or detoxification genes. More generally, transfer being the highest within kin, together with the observation that plasmids are not at fixation within lineages in the field [16] suggests that most plasmid dynamics might actually occur not between lineages (the events most easily detected) but within lineages, leading to specific coevolution of plasmids with specific host lineages despite recurring dynamics of plasmid transfer and loss.


Alam, M.T., T.D. Read, R.A. Petit III, S. Boyle-Vavra, L.G. Miller, S.J. Eells, R.S. Daum, M.Z. David. 2015. “Transmission and Microevolution of USA300 MRSA in U.S. Households: Evidence from Whole-Genome Sequencing.” mBio 6 (2): e00054–15. doi:10.1128/mBio.00054-15.

American Society for Microbiology. 2015. “MRSA Can Linger in Homes, Spreading among Its Inhabitants.” ScienceDaily.

Fox, J., ed. 2015. The Threat of MRSA. American Academy of Microbiology. Washington, DC.

Gillen, A. L. 2015. The Genesis of Germs: Disease and the Coming Plagues in a Fallen World. Green Forest, Arkansas: Master Books.

Exploitation of Bacillus subtilis as a robust workhorse for production of heterologous proteins and beyond

Bacillus subtilis, belonging to the type species of Bacillus, is a type of soil-derived, low %G+C, endospore-forming Gram-positive bacterium. After the discovery of B. subtilis 168 that displayed natural competence, this bacterium has been intensively considered to be an ideal model organism and a robust host to study several basic mechanisms, such as metabolism, gene regulation, bacterial differentiation, and application for industrial purposes, such as heterologous protein expression and the overproduction of an array of bioactive molecules. Since the first report of heterologous overproduction of recombinant proteins in this strain, the bulk production of a multitude of valuable enzymes, especially industrial enzymes, has been performed on a relatively large scale. Since B. subtilis can non-specifically secrete recombinant proteins using various signal peptides, it has tremendous advantages over Gram-negative bacterial hosts. Along with the report of the complete genome sequence of B. subtilis, a number of genetic tools, including diverse types of plasmids, bacterial promoters, regulatory elements, and signal peptides, have been developed and characterized. These novel genetic elements tremendously accelerated genetic engineering in B. subtilis recombinant systems. In addition, with the development of several complex gene expression systems, B. subtilis has performed a number of more complex functions. This ability enables it to be a substantial chassis in synthetic biology rather than just a workhorse for the overproduction of recombinant proteins. In this review, we review the progress in the development of B. subtilis as a universal platform to overproduce heterologous diverse high-value enzymes. This progress has occurred from the development of biological parts, including the characterization and utilization of native promoters, the fabrication of synthetic promoters and regulatory elements, and the assembly and optimization of genetic systems. Some important industrial enzymes that have been produced in large quantities in this host are also summarized in this review. Furthermore, the ability of B. subtilis to serve as a cellular tool was also briefly recapitulated and reviewed.

This is a preview of subscription content, access via your institution.


Genetic elements called transposons (transposable elements), or &ldquojumping genes,&rdquo are molecules of DNA that include special inverted repeat sequences at their ends and a gene encoding the enzyme transposase (Figure (PageIndex<5>)). Transposons allow the entire sequence to independently excise from one location in a DNA molecule and integrate into the DNA elsewhere through a process called transposition. Transposons were originally discovered in maize (corn) by American geneticist Barbara McClintock (1902&ndash1992) in the 1940s. Transposons have since been found in all types of organisms, both prokaryotes and eukaryotes. Thus, unlike the three previous mechanisms discussed, transposition is not prokaryote-specific. Most transposons are nonreplicative, meaning they move in a &ldquocut-and-paste&rdquo fashion. Some may be replicative, however, retaining their location in the DNA while making a copy to be inserted elsewhere (&ldquocopy and paste&rdquo). Because transposons can move within a DNA molecule, from one DNA molecule to another, or even from one cell to another, they have the ability to introduce genetic diversity. Movement within the same DNA molecule can alter phenotype by inactivating or activating a gene.

Transposons may also carry with them additional genes, moving these genes from one location to another with them. For example, bacterial transposons can relocate antibiotic resistance genes, moving them from chromosomes to plasmids. This mechanism has been shown to be responsible for the co-localization of multiple antibiotic resistance genes on a single R plasmid in Shigella strains causing bacterial dysentery. Such an R plasmid can then be easily transferred among a bacterial population through the process of conjugation.

Figure (PageIndex<5>): Transposons are segments of DNA that have the ability to move from one location to another because they code for the enzyme transposase. In this example, a nonreplicative transposon has disrupted gene B. The consequence of that the transcription of gene B may now have been interrupted.

What are two ways a transposon can affect the phenotype of a cell it moves to?

Table (PageIndex<1>) summarizes the processes discussed in this section.

Table (PageIndex<1>): Summary of Mechanisms of Genetic Diversity in Prokaryotes
Term Definition
Conjugation Transfer of DNA through direct contact using a conjugation pilus
Transduction Mechanism of horizontal gene transfer in bacteria in which genes are transferred through viral infection
Transformation Mechanism of horizontal gene transfer in which naked environmental DNA is taken up by a bacterial cell
Transposition Process whereby DNA independently excises from one location in a DNA molecule and integrates elsewhere

Clinical Focus: Resolution

Within 24 hours, the results of the diagnostic test analysis of Alex&rsquos stool sample revealed that it was positive for heat-labile enterotoxin (LT), heat-stabile enterotoxin (ST), and colonization factor (CF), confirming the hospital physician&rsquos suspicion of ETEC. During a follow-up with Alex&rsquos family physician, this physician noted that Alex&rsquos symptoms were not resolving quickly and he was experiencing discomfort that was preventing him from returning to classes. The family physician prescribed Alex a course of ciprofloxacin to resolve his symptoms. Fortunately, the ciprofloxacin resolved Alex&rsquos symptoms within a few days.

Alex likely got his infection from ingesting contaminated food or water. Emerging industrialized countries like Mexico are still developing sanitation practices that prevent the contamination of water with fecal material. Travelers in such countries should avoid the ingestion of undercooked foods, especially meats, seafood, vegetables, and unpasteurized dairy products. They should also avoid use of water that has not been treated this includes drinking water, ice cubes, and even water used for brushing teeth. Using bottled water for these purposes is a good alternative. Good hygiene (handwashing) can also aid the prevention of an ETEC infection. Alex had not been careful about his food or water consumption, which led to his illness.

Alex&rsquos symptoms were very similar to those of cholera, caused by the gram-negative bacterium Vibrio cholerae, which also produces a toxin similar to ST and LT. At some point in the evolutionary history of ETEC, a nonpathogenic strain of E. coli similar to those typically found in the gut may have acquired the genes encoding the ST and LT toxins from V. cholerae. The fact that the genes encoding those toxins are encoded on extrachromosomal plasmids in ETEC supports the idea that these genes were acquired by E. coli and are likely maintained in bacterial populations through horizontal gene transfer.

Watch the video: Patagonia: The Paradox of an Eco-Conscious Company (June 2022).


  1. Haig

    You are absolutely right. There is something in it also to me it seems a good thought. I agree with you.

  2. Medwyn

    I didn't understand everything.

  3. Iasius

    That does not concern you!

  4. Heath

    Sorry for interfering, there is a suggestion that we should take a different route.

  5. Lapu

    Well done, this very good idea is just about

  6. Pekar

    In it all grace!

  7. Omet

    Authoritative point of view, curious ..

  8. Ruairidh

    I find that you are not right. I can prove it. Write in PM, we will talk.

Write a message