A PROCESS FOR PRODUCING A CARRIER ORGANISM CONTAINING A HETEROLOGOUS METAGENOMIC DN A

This application claims priority of U.S. Provisional Application No. 61 /299, 157, filed January 28, 2010, the content of which is hereby incorporated by reference.

Throughout this application, certain publications are referenced. Full citations for these publications may be found immediately preceding the claims. The disclosures of these publications in their entireties are hereby incorporated by reference into this application in order to describe more fully the state-of-the art to which this invention relates.

Background Of The Invention

From the 1930's until the 1980s, scientists in pharmaceutical companies and academic laboratories screened millions of soil samples from around the world looking for pharmaceutically active compounds produced by bacteria in the soil samples. The effort was a great success. Many new antibacterial agents were discovered along with anticancer drugs like doxorubicin and mithramycin and immunomodulators like rapamycin and cyclosporine. These naturally occurring compounds in turn gave rise to numerous semisynthetic derivatives and totally new synthetic compounds based on these structures found in nature. The success of these soil screening programs accounts, in part, for the estimate that 60-70% of the pharmaceutical agents currently in use are natural products themselves or are based on natural products.

The success of the soil screening programs is even more impressive when one learns that only about 1 % of the bacteria in soil samples will grow under the conditions normally employed in laboratories (Torsvik et a!. 1990). So, for all those years and all those samples, scientists were looking at only 1% of the activity they contained. Recently, various approaches have been taken to gain access to the unculturable majority. In the most straightforward approach, trial and error can eventually lead to the proper combination of growth media and conditions needed to cultivate a newly identified bacterial strain (Davis et al. 2005; Stevenson et al. 2004). In another technological development it was demonstrated that isolating individual microbes in microdroplets within a gel emulsion allowed for the growth of a variety of bacterial species that otherwise could not be culture en masse (Zengler et al. 2002). Unfortunately, neither of these methodologies is suitable for the large scale identification of novel compounds in mammalian cell assays. The first method is time consuming and limited in scope while the latter method does not provide a means for easy incorporation into mammalian cell culture assays. Additionally, both methods lessen the chances of identifying enzymatic activities and/or natural compounds that are only transiently produced in response to specific stimuli.

In recent years some efforts have been undertaken to isolate unique enzymes from metagenomic-derived DNA libraries. Far less emphasis has been placed on isolating large fragments of chromosomal DNA encompassing metabolic pathways. These large fragments of chromosomal DNA code for proteins needed to synthesize unique drugs (e.g. an anticancer compound). In another attempt to overcome culturing issues, plasmid or BACmid libraries with large inserts have been generated directly from metagenomic samples taken from soil, seawater, etc. Standard screening methods of such libraries would involve making extracts from individual bacterial colonies and then using those extracts in one-extract-per-well cell based assays. Though one-extract-per-well and related methods have merit, there are certain limitations. For one, serum and other proteins used in cell growth media could potentially interfere with drug delivery. Also, issues such as storing extract, degradation of drug during freeze-thaw, toxicity of extract to cells, and light sensitivity of drugs also come into play. Further, existing methods don't allow for realistic screening in model organisms such as nematodes (worms) and mice without the possible use of expensive robotics.

Microfluidic chip technology has proven to be a very versatile and a reliable means to miniaturize many standard technologies available to researchers thus earning it the name "lab-on-a-chip" (Chovan et al. 2002; Weigl et al. 2003). Recent studies have demonstrated that conditions within microdroplets can be suitable for the survival and growth of microorganisms such as bacteria (Huebner et al. 2007) and yeast (Luo et al. 2006). The incorporation of gas permeable fluorocarbonated oil as well as biocompatible surfactants has allowed for the survival and growth of mammalian cell types such as 2C6 hybridoma cells and the Jurkat T cell leukemia cell line (Clausell-Tormos et al. 2008; Koster et al. 2008). The mixing of oil and cell media at determined flow rates through channels etched into glass or polydimethvlsiloxane (PDMS) chips has made it is possible to generate steady streams of microdroplets of a predetermined size and, once formed, to manipulate these droplets. As such, an almost endless number of existing experimental conditions can be adapted to microfluidic droplet format. Any minor deviations in standard on-the-bench protocols have been more than offset by the fact that researchers utilizing microfluidic chips can perform hundreds of thousands or even millions of experiments an hour.

Proof-of-concept experiments have found suitable microfluidic chip designs and reagents that would allow for cell and organism based drug screening. To date, microfluidic chip technology has not been used for drug screening in mammalian cells. It has been suggested that traditional high throughput screening (HTS) could be coupled with microdroplet screening if small molecule libraries with diffusion resistant oils were utilized (Courtois et al. 2009; Huebner et al. 2008). This would certainly miniaturize currently available HTS assays but would confine researchers to the limitations inherent to small molecules produced by combinatorial chemistry. Though natural product based mammalian drug screens have been suggested ( oster et al. 2008), no satisfactory methodology has been offered up that would allow for efficient screening of the vast number of chemical compounds found in nature.

Summary Of The Invention

This invention provides an organism containing a heterologous metagenomic DNA.

This invention provides a process for producing a carrier organism containing a heterologous metagenomic DNA, the gene product of which heterologous metagenomic DNA results when present in a cell from a carrier organism other than the organism which is the source of the heterologous metagenomic DNA, which process comprises:

i) obtaining the heterologous metagenomic DNA from the source organism; and

ii) introducing the metagenomic DNA into a carrier organism.

This invention provides a process for recovering a carrier organism containing a heterologous metagenomic DNA, the gene product of which heterologous metagenomic DNA results when present in a cell from a carrier organism other than the organism which is the source of the heterologous metagenomic DNA, which process comprises:

i) obtaining the heterologous metagenomic DNA from the source organism;

ii) introducing the metagenomic DNA into a carrier organism; iii) adding the carrier organism from step (ii) to one or more test subjects;

iv) assaying the one or more test subjects to identify test subjects which exhibit the desired phenotype;

v) separating the test subjects exhibiting the desired phenotype indentified in step (iv); and

vi) recovering from the test subjects exhibiting the desired phenotype that are separated in step (v) the carrier organism containing a heterologous metagenomic DNA.

This invention provides a process for delivering a gene product into cells, the presence of which gene product results in the cells exhibiting a desired phenotype which comprises:

i) obtaining a carrier organism containing metagenomic DNA that encodes a gene product that results in the cells of a test subject exhibiting a desired phenotype; and ii) contacting the carrier organism from step (ii) with the cells.

This invention provides a screening method for identifying a metagenomic DNA, the gene product of which metagenomic DNA results in expression of a desired phenotype of a cell from an organism other than the source of the metagenomic DNA, which method comprises:

i) obtaining a carrier organism containing metagenomic DNA from a heterlogous source;

ii) adding the carrier organism from step (i) to one or more test subjects;

iii) assaying one or more of such test subjects to identify test subjects which exhibit the desired phenotype;

iv) separating the test subjects exhibiting the desired phenotype indentified in step (iii);

v) recovering from the test subjects exhibiting the desired phenotype that are separated in step (iv) the carrier organism containing metagenomic DNA from a heterologous source; and vi) identifying the metagenomic DNA contained in the carrier organism.

Brief Description of the Figures

Figure 1: Modified bacteria expressing metagenomic DNA invade cancer cells and export novel metabolites or release novel metabolites when lysed.

Figure 2A, 2B, 2C: Metagenomic drug screening with invasive bacteria. A) top panel - Environmental sample (e.g. soil) is collected, microbes are isolated and metagenomic DNA isolated. DNA is then introduced to invasive bacteria. Bottom panel - invasive bacteria producing novel metabolite are grown and introduced to cancer cell, B) invade cancer cell, C) are lysed, allowing D) the release of novel metabolites which result in E) activation of a reporter gene for cell sorting. The remaining intact bacteria are then harvested from the cancer cell so that their metagenomic DNA can be sequenced and their metabolite production analyzed in detail.

Figure 3: Screening via microfluidics. A microfluidics chip is designed with a merger element that mechanically fuses two microdroplets, one droplet containing a clonal population of invasive bacteria producing metagenomic metabolites and a second droplet containing a clonal population of cancer cells. Fused drops are guided out of the chip and disrupted to release infected cancer cells in the presence of gentamycin to lyse extracellular bacteria.

Figure 4: Invasive bacteria mediated delivery of anticancer drug violacein. A) E. coli rendered invasive (left panel) and invasive bacteria expressing the violacein operon (right panel), as evidenced by the deep violet color of the colonies. B) Cell death brought about by invasive bacteria producing violacein. 24 hours after treatment, HeLa S3 cancer cells treated with invasive bacteria producing violacein (bottom right panel) have mostly undergone cell death. Control cells expressing the invasin protein alone (upper right panel) or the violacein operon alone (lower left panel) did not affect the HeLa S3 cells. Figure 5: Invasive violacein producing bacteria kill via apoptosis. HeLa S3 cells were infected with bacteria producing violacein and control cells at a MOI of 25. At 6 hours post infection live cells were clearly positive for Annex in V-FITC staining, (lower right panel, white arrows). The widespread appearance of vacuoles occurred only with invasive cells infected with violacein producing bacteria (upper right panel, yellow arrows).

Figure 6: AIEC strain LF82 expressing invasin shows widespread infection and intracellular replication in J774A.1 cells. LF82 bacteria were engineered to carry two plasmids, one expressing invasin and a second plasmid expressing red fluorescent protein (RFP). After one hour of infection (top panels) each J774A.1 cell had between 2-5 RFP positive LF82 bacteria. At 15 hours post- infection (bottom panels) many J774A.1 cells experienced an intracellular replication of the internalized LF82 RFP cells.

Figure 7A, 7B: (A) Bacterial delivery of a drug payload. Invasive E. coli producing a natural product enter the cytoplasm of mammalian cells and are lysed in phagosomes. The released natural product (red pentagon) is then fee to bind to its target protein (green) (B) Vector TRIP 3.0. The promoter P- Amp drives dual expression of invasin and lysteriolysin. Invasin expression allows for internalization by mammalian cells expressing betal-integrin and listeriolysin disrupts the integrity of phagocytic vesicles for efficient release of the contents from lysed bacteria into the cytosol.

Figure 8A, 8B: (A) Violacein expression in invasive Ecoli results in dark, pigmented colonies (right panel) that allowed for trackable expression. (B) Four hours post-treatment of HeLa S3 cells there is no effect from bacteria expressing empty vector (upper left), vioA-E operon (upper right), invasin (lower left) but internalization of vioA-E (lower right) results in the appearance of multiple vacuoles (red arrows) in many cells.

Figure 9: Invasive violacein producing bacteria kill via apoptosis. HeLa S3 cells were infected with bacteria producing violacein and control cells at a MOI of 25. At 6 hours post infection live cells were clearly positive for Annexin V-FITC staining, (lower right panel, white arrows). The widespread appearance of vacuoles occurred only with invasive cells infected with violacein producing bacteria (upper right panel, yellow arrows).

Figure 10: T-junction chip. Injection of a carrier oil and cell media at predetermined rates produces nanoliter sized droplets which can be further manipulated.

Figure 11: Formation of droplets with cancer cell culture media. Dulbecco's Modified

Eagle Medium (DMEM) supplemented with 10% fetal bovine serum efficiently formed droplets in Hexadecane containing 1% Span 80 surfactant.

Figure 12: A Novel Metagenomic Screening Method For Drug Discovery

Figure 13: Invasive Bacteria Payload

Figure 14: Bacterial Lysis Over Time

Figure 15: Improved Metagenomic Drug Mining

Figure 16: Identification of Unique Metabolic Pathways Involving Recovery of bacteria and isolateion of B ACmid.

Figure 17: Identification and Incubation of Cells with Antibiotics

Figure 18: In Vitro and in vivo applications

Figure 19: Test Case using Violacein

Figure 20: Bacterial Delivery of Violacein in Vero cells

Figure 21: Bacterial Delivery of Violacein in HeLa S3 cells

Figure 22: Invasive Bacteria Deliver Payload Oetailed Pescrtptioii of the Invention

For convenience, before further description of the present invention, certain terms employed in the specification, examples and appended claims are defined here.

Metagenomic DNA is DNA comprised of the genomic DNA of many microorganisms in an environment, many of which cannot be cultured using standard methods.

Apoptosis means an orderly process of programmed cell-death (PCD) which is characterized by a defined set of events that result in the safe disposal of unwanted cells.

Heterologous DNA means DNA that is derived from an organism other than the organism in which it is present.

This invention provides an organism containing a heterologous metagenomic DNA.

In an embodiment of the above organism, the organism is capable of invading a second organism.

In an embodiment of the above organism, the organism contains a bacterial invasion gene. In another embodiment, the organism contains a bacterial invasion gene cluster. In another embodiment, the organism contains a bacterial invasion gene operon. In a further embodiment, the bacterial invasion gene is a Salmonella gene. In a further embodiment, the bacterial invasion gene is a Y. pseudotuberculosis gene. In a further embodiment, the bacterial invasion gene is a M. tuberculosis gene. In a further embodiment, the organism expresses viral cell proteins which allow for invasion. In a further embodiment, the organism expresses mammalian cell proteins which allow for invasion.

In an embodiment of any of the above organisms, the metagenomic DNA is 5kbp or more in length.

In an embodiment of any of the above organisms, the organism is a bacteria. In a further embodiment, the bacteria is an attenuated E. coli. In a further embodiment, the bacteria is of the genus Bacillus, In a further embodiment, the bacteria is of the genus Streptomyces. In a further embodiment, the bacteria is of the genus Actinomyces. In a further embodiment, the bacteria is Agrobacterium tumefaciens. In a further embodiment, the bacteria is Burkholderia graminis. In a further embodiment, the bacteria is Caulobacter vibrioides. In a further embodiment, the bacteria is Pseudomonas putids. In a further embodiment, the bacteria is Ralstonia metallidurans.

In an embodiment of any of the above organisms, the organism is a fungal cell.

In an embodiment of any of the above organisms, the organism is a parasitic protozoa.

In an embodiment of any of the above organisms, the expression of the metagenomic DNA results in the organism exhibiting a desired phenotype in a cell from an organism other than the source of the heterologous metagenomic DNA.

In an embodiment of any of the above organisms, when the organism is in a nematode the expression of the heterologous metagenomic DNA results in the nematode exhibiting a desired phenotype. In another embodiment, when the organism is in a mammalian cell the expression of the heterologous metagenomic DNA results in the mammalian cell exhibiting a desired phenotype. In another embodiment, when the organism is in an amoeba the expression of the heterologous metagenomic DNA results in the amoeba exhibiting a desired phenotype. In another embodiment, when the organism is in a mouse the expression of the heterologous metagenomic DNA results in the mouse exhibiting a desired phenotype. In another embodiment, when the organism is in a mouse cell the expression of the heterologous metagenomic DNA results in the mouse cell exhibiting a desired phenotype. In another embodiment, when the organism is in a mammalian cell the expression of the heterologous metagenomic DNA results in the mammalian exhibiting a desired phenotype. In another embodiment, when the organism is in a human cell the expression of the heterologous metagenomic DNA results in the human cell exhibiting a desired phenotype. In another embodiment, when the organism is in a cell that is capable of ingesting the organism the expression of the heterologous metagenomic DNA results in the cell that is capable of ingesting the organism exhibiting a desired phenotype. In another embodiment, when the organism is in a cell that is capable of being infected by the organism the expression of the heterologous metagenomic D A results in the cell that - Il ls capable of being infected by the organism exhibiting a desired phenotype. In another embodiment, when the organism is in a unicellular eukaryote the expression of the heterologous metagenomic DNA results in the unicellular eukaryote exhibiting a desired phenotype. In another embodiment, when the organism is in a muulticellular eukaryote the expression of the heterologous metagenomic DNA results in the multicellular eukaryote exhibiting a desired phenotype.

In an embodiment of any of the above organisms, the desired phenotype can be assessed in an assay. In another embodiment, the desired phenotype can be assessed by use of a stain. In another embodiment, the desired phenotype can be assessed by the activation of a reporter gene. In another embodiment, the desired phenotype can be assessed in an activity assay. In another embodiment, the desired phenotype is cell survival. In another embodiment, the desired phenotype is differentiation. In another embodiment, the desired phenotype is dedifferentiation. In another embodiment, the desired phenotype is cell migration. In another embodiment, the desired phenotype is DNA repair. In another embodiment, the desired phenotype is apoptosis or cell death. In another embodiment, the desired phenotype is cell growth. In another embodiment, the desired phenotype is cell multiplication. In another embodiment, the desired phenotype is cell multiplication, such as increasing the number of liver cells in a dish instead of obtaining stem cells.

In an embodiment of any of the above organisms, the desired phenotype is the expression of a reporter gene. In a further embodiment, the reporter gene produces a colored product. In a further embodiment, the reporter gene encodes red, green or yellow fluorescent protein.

In an embodiment of any of the above organisms, the heterologous metagenomic DNA is present in a soil sample. In another embodiment, the heterologous metagenomic DNA is present in a water sample. In another embodiment, the heterologous metagenomic DNA is present in a library of DNA from mixed sources. In another embodiment, the heterologous metagenomic DNA is present on the surface of a plant. In another embodiment, the heterologous metagenomic DNA is present in a microbial community associated with an organism.. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that is present on the surface of human skin. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that is present in the alimentary tract of a mosquito. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that is present in the alimentary tract of a mouse. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that is present in the alimentary tract of a caterpillar. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that has a symbiotic relationship with an organism, such as a bacteria that produces compounds that are toxic in high doses to provide defense to its host, such as a marine sponge.

In an embodiment of any of the above organisms, the heterologous metagenomic DNA is between 500bp and 250kbp in length.

In an embodiment of any of the above organisms, the carrier organism contains heterologous metagenomic DNA encoding a single metabolic pathway. In another embodiment, the single metabolic pathway produces a known drug. In a further embodiment, the DNA encoding a single metabolic pathway is prone to mutation. In another embodiment, the carrier organism is treated with a mutagen.

In an embodiment of any of the above organisms, the organism has been modified to allow endosomal escape via coating with recombinant Listeriolysin O.

This invention provides a process for recovering a carrier organism containing a heterologous metagenomic DNA, the gene product of which heterologous metagenomic DNA results when present in a cell from a carrier organism other than the organism which is the source of the heterologous metagenomic DNA, which process comprises:

i) obtaining the heterologous metagenomic DNA from the source organism;

ii) introducing the metagenomic DNA into a carrier organism;

iii) adding the carrier organism from step (ii) to one or more test subjects;

iv) assaying the one or more test subjects to identify test subjects which exhibit the desired phenotype;

v) separating the test subjects exhibiting the desired phenotype indentified in step (iv); and vi) recovering from the test subjects exhibiting the desired phenoiype that are separated in step (v) the carrier organism containing a heterologous metagenomic DNA.

In one embodiment of the above process, the carrier organism of step (ii) is grown on an alginate bead in the presence of the one or more test subjects for a time sufficient for the carrier organism to release a heterologous metagenomic DNA gene product.

In one embodiment of any of the above processes, the carrier organism is capable of invading a second organism.

In one embodiment of any of the above processes, the process further comprises incubating a mixture of the carrier organism and one or more test subjects from step (iii) for a time sufficient for the carrier organism of step (ii) to invade the test subjects of step (iii).

In one embodiment of any of the above processes, the process further comprises incubating the test subjects that were invaded by the carrier organism.

In one embodiment of any of the above processes, the process further comprises adding antibiotic to kill any carrier organisms which have not invaded the test subjects after the test subjects were incubated.

In one embodiment of any of the above processes, the process further comprises the steps of injecting the carrier organism of step (iii) into a microfluidics chip bacterial drop generator circuit set to generate drops encapsulating 1 or more bacteria cells, collecting drops containing the carrier organism, incubating the carrier organism containing drops for a time sufficient for the bacteria to saturate the drops, and reinjecting drops containing transformed bacteria into the microfluidics chip generator before adding the carrier organism to one or more test subjects in step (iii).

In one embodiment of any of the above processes, the one or more test subjects of step (iii) are formed in drops. In one embodiment of any of the above processes, the carrier organism is added to the one or more test subjects in step (iii) by injecting drops containing the one or more test subjects with the microfluidics chip generator such that each drop containing the one or more test subjects will be paired with a single drop containing the carrier organism.

In one embodiment of any of the above processes, an individual pair of drops containing one or more mammalian test subject cells and the carrier organism are merged into a single merged drop.

In one embodiment of any of the above processes, the process further comprises incubating the single merged drop for a time sufficient for the carrier organism of step to invade the one or more test subjects.

In one embodiment of any of the above processes, the process further comprises incubating the one or more test subjects that were invaded by the carrier organism.

In one embodiment of any of the above processes, the process further comprises adding antibiotic to kill any carrier organisms which have not invaded the test subjects after the test subjects were incubated.

In one embodiment of any of the above processes, each individual pair of drops formed are merged by applying an electrical field to the microfulidics chip bacterial drop generator.

In one embodiment of any of the above processes, each individual pair of drops formed are mechanically merged.

In one embodiment of any of the above processes, the drops containing one or more test cells contains a sphere coated with the test cells.

In one embodiment of any of the above processes, the sphere is made of Dextran. In a further embodiment, the sphere is made of gelatin. In a further embodiment, the sphere is made of alginate. In a further embodiment, the sphere is made of an alginate derivative. In a further embodiment, the sphere is made ofan alginate mixed with an additional agent. In a further embodiment, the sphere is made of, the additional agent is a cell binding peptide. In a further embodiment, the additional agent is a magnetic material. This invention provides a process for producing a carrier organism containing a heterologous metagenomic DNA, the gene product of which heterologous metagenomic DNA results when present in a cell from a carrier organism other than the organism which is the source of the heterologous metagenomic DNA, which process comprises:

i) obtaining the heterologous metagenomic DNA from the source organism; and

ii) introducing the metagenomic DNA into a carrier organism.

This invention provides a process for delivering a gene product into cells, the presence of which gene product results in the cells exhibiting a desired phenotype which comprises:

i) obtaining a carrier organism containing metagenomic DNA that encodes a gene product that results in the cells of a test subject exhibiting a desired phenotype; and

ii) contacting the carrier organism from step (ii) with the cells.

In one embodiment of the above process the metagenomic DNA containing carrier organism is administered orally. In another embodiment, the metagenomic DNA containing carrier organism is administered ballistically. In a further embodiment, the metagenomic DNA containing carrier organism is administered ballistically with a helium powered gene gun.

In one embodiment of any of the above processes, the organism is capable of invading a second organism.

In one embodiment of any of the above processes, the organism contains a bacterial invasion gene. In another embodiment, the organism contains a bacterial invasion gene cluster. In another embodiment, the organism contains a bacterial invasion gene operon. In a further embodiment, the bacterial invasion gene is a Salmonella gene. In a further embodiment, the bacterial invasion gene is a Y. pseudotuberculosis gene. In a further embodiment, the bacterial invasion gene is a M. tuberculosis gene. In a further embodiment, the organism expresses viral cell proteins which allow for invasion. In a further embodiment, the organism expresses mammalian cell proteins which allow for invasion.

In one embodiment of any of the above processes, the metagenomic DNA is 5kbp or more in length.

In one embodiment of any of the above processes, the organism is a bacteria. In a further embodiment, the bacteria is an attenuated E. coli. In a further embodiment, the bacteria is of the genus Bacillus. In a further embodiment, the bacteria is of the genus Streptomyces. In a further embodiment, the bacteria is of the genus Actinomyces. In a further embodiment, the bacteria is Agrobacterium tumefaciens. In a further embodiment, the bacteria is Burkholderia graminis. In a further embodiment, the bacteria is Caulobacter vibrioides. In a further embodiment, the bacteria is Pse domonas put ids. In a further embodiment, the bacteria is Ralstonia metallidurans.

In one embodiment of any of the above processes, the test subject is a nematode. In another embodiment, the test subject is a mammalian cell. In another embodiment, the test subject is an amoeba. In another embodiment, the test subject is a mouse. In another embodiment, the test subject is a mouse cell. In another embodiment, the test subject is a human cell. In another embodiment, the test subject is derived from a mammalian cell. In another embodiment, the test subject is capable of ingesting the carrier organism. In another embodiment, the test subject is capable of being infected by the carrier organism. In another embodiment, the test subject is a unicellular eukaryote. In another embodiment, the test subject is a multicullular eukaryote.

In one embodiment of any of the above processes, the organism is a fungal cell.

In one embodiment of any of the above processes, the organism is a parasitic protozoa.

In one embodiment of any of the above processes, the expression of the metagenomic DNA results in the organism exhibiting a desired phenotype in a cell from an organism other than the source of the heterologous metagenomic DNA. In one embodtment of any of the above processes, when the organism is in a nematode the expression of the heterologous metagenomic DNA results in the nematode exhibiting a de ired phenotype. In another embodiment, when the organism is in a mammalian cell the expression of the heterologous metagenomic DNA results in the mammalian cell exhibiting a desired phenotype. In another embodiment, when the organism is in an amoeba the expression of the heterologous metagenomic DNA results in the amoeba exhibiting a desired phenotype. In another embodiment, when the organism is in a mouse the expression of the heterologous metagenomic DNA results in the mouse exhibiting a desired phenotype. In another embodiment, when the organism is in a mouse cell the expression of the heterologous metagenomic DNA results in the mouse cell exhibiting a desired phenotype. In another embodiment, when the organism is in a mammalian cell the expression of the heterologous metagenomic DNA results in the mammalian exhibiting a desired phenotype. In another embodiment, when the organism is in a human cell the expression of the heterologous metagenomic DNA results in the human cell exhibiting a desired phenotype. In another embodiment, when the organism is in a cell that is capable of ingesting the organism the expression of the heterologous metagenomic DNA results in the cell that is capable of ingesting the organism exhibiting a desired phenotype. In another embodiment, when the organism is in a cell that is capable of being infected by the organism the expression of the heterologous metagenomic DNA results in the cell that is capable of being infected by the organism exhibiting a desired phenotype. In another embodiment, when the organism is in a unicellular eukaryote the expression of the heterologous metagenomic DNA results in the unicellular eukaryote exhibiting a desired phenotype. In another embodiment, when the organism is in a muulticellular eukaryote the expression of the heterologous metagenomic DNA results in the multicellular eukaryote exhibiting a desired phenotype.

In one embodiment of any of the above processes, the desired phenotype can be assessed in an assay. In another embodiment, the desired phenotype can be assessed by use of a stain. In another embodiment, the desired phenotype can be assessed by the activation of a reporter gene. In another embodiment, the desired phenotype can be assessed in an activity assay. In another embodiment, the desired phenotype is cell survival. In another embodiment, the desired phenotype is differentiation. In another embodiment, the desired phenotype is dedifferentiation. In another embodiment, the desired phenotype is cell migration. In another embodiment, the desired phenotype is DNA repair. In another - X8 ~ embodiment, the desired phenotype is apoptosis or cell death. In another embodiment, the desired phenotype is cell growth. In another embodiment, the desired phenotype is cell multiplication. In another embodiment, the desired phenotype is cell multiplication, such as increasing the number of liver cells in a dish instead of obtaining stem cells.

In one embodiment of any of the above processes, the desired phenotype is the expression of a reporter gene. In a further embodiment, the reporter gene produces a colored product. In a further embodiment, the reporter gene encodes red, green or yellow fluorescent protein.

In one embodiment of any of the above processes, the heterologous metagenomic DNA is present in a soil sample. In another embodiment, the heterologous metagenomic DNA is present in a water sample. In another embodiment, the heterologous metagenomic DNA is present in a library of DNA from mixed sources. In another embodiment, the heterologous metagenomic DNA is present on the surface of a plant. In another embodiment, the heterologous metagenomic DNA is present in a microbial community associated with an organism.. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that is present on the surface of human skin. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that is present in the alimentary tract of a mosquito. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that is present in the alimentary tract of a mouse. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that is present in the alimentary tract of a caterpillar. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that has a symbiotic relationship with an organism, such as a bacteria that produces compounds that are toxic in high doses to provide defense to its host, such as a marine sponge.

In one embodiment of any of the above processes, a nutrient or precursor drug is added to at least one of the drops. In a further embodiment, the nutrient or precursor drug may be absorbed and modified by the carrier organism. In one embodiment, the precursor drug is the basic skeletal structure of aspirin. In another embodiment, the nutrient or precursor drug is, for example, tyrosine, the addition of which promotes the production of gene products such as violacein and other bacterial metabolites which have tyrosine as a precursor.

In one embodiment of any of the above processes, the drops are 5 or more microns in diameter. In another embodiment, the drops are about 30 microns to about 100 microns in diameter.

In one embodiment of any of the above processes, an oil is used to generate the drops. In a further embodiment, a hydrocarbon oil is used to generate the drops. In a further embodiment, a mineral oil is used to generate the drops. In a further embodiment, a fluorinated oil is used to generate the drops. In another embodiment, the drops contain a surfactant.

In one embodiment of any of the above processes, the heterologous metagenomic DNA is between 500bp and 250kbp in length.

In an embodiment of any of the above processes, the carrier organism contains heterologous metagenomic DNA encoding a single metabolic pathway. In another embodiment, the single metabolic pathway produces a known drug. In a further embodiment, the DNA encoding a single metabolic pathway is prone to mutation. In another embodiment, the carrier organism is treated with a mutagen.

In one embodiment of any of the above processes, the organism has been modified to allow endosomal escape via coating with recombinant Listeriolysin O.

This invention provides a screening method for identifying a metagenomic DNA, the gene product of which metagenomic DNA results in expression of a desired phenotype of a cell from an organism other than the source of the metagenomic DNA, which method comprises:

i) obtaining a carrier organism containing metagenomic DNA from a heterlogous source;

ii) adding the carrier organism from step (i) to one or more test subjects; iii) assaying one or more of such test subjects to identify test subjects which exhibit the desired phenotype;

iv) separating the test subjects exhibiting the desired phenotype indentified in step (iii);

v) recovering from the test subjects exhibiting the desired phenotype that are separated in step (iv) the carrier organism containing metagenomic DNA from a heterologous source; and vi) identifying the metagenomic DNA contained in the carrier organism.

In an embodiment of the above method, the carrier organism of step (ii) is grown on an alginate bead in the presence of the one or more test subjects for a time sufficient for the carrier organism to release a metagenomic DNA gene product.

In an embodiment of any of the above methods, the carrier organism is capable of invasion.

In an embodiment of any of the above methods, the method further comprises incubating a mixture of the carrier organism and one or more test subjects from step (ii) for a time sufficient for the carrier organism of step (i) to invade the test subjects of step (ii).

In an embodiment of any of the above methods, the method further comprises incubating the test subjects that were invaded by the carrier organism.

In an embodiment of any of the above methods, the method further comprises adding antibiotic to kill any carrier organisms which have not invaded the test subjects after the test subjects were incubated.

In an embodiment of any of the above methods, the method further comprises the steps of injecting the carrier organism of step (i) into a microfluidics chip bacterial drop generator circuit set to generate drops encapsulating 1 or more bacteria cells, collecting drops containing the carrier organism, incubating the carrier organism containing drops for a time sufficient for the bacteria to saturate the drops, and reinjecting drops containing transformed bacteria into the microfluidics chip generator before adding the carrier organism to one or more test subjects in step (ii).

In an embodiment of any of the above methods, wherein the one or more test subjects of step (ii) are formed in drops.

In an embodiment of any of the above methods, wherein the carrier organism is added to the one or more test cells in step (ii) by injecting drops containing the one or more test cells with the microfluidics chip generator such that each drop containing the one or more test cells will be paired with a single drop containing the carrier organism.

In an embodiment of any of the above methods, wherein an individual pair of drops containing the one or more mammalian test cells and the carrier organism are merged into a single merged drop.

In an embodiment of any of the above methods, the method further comprises incubating the single merged drop for a time sufficient for the carrier organism of step to invade the one or more test subject.

In an embodiment of any of the above methods, the method further comprises incubating the one or more test subjects that were invaded by the carrier organism.

In an embodiment of any of the above methods, the method further comprises adding antibiotic to kill any carrier organisms which have not invaded the test subjects after the test subjects were incubated.

In an embodiment of any of the above methods, wherein each individual pair of drops formed are merged by applying an electrical field to the microfulidics chip bacterial drop generator.

In an embodiment of any of the above methods, wherein each individual pair of drops formed are mechanically merged.

In one embodiment of any of the above methods, a nutrient or precursor drug is added to at least one of the drops. In a further embodiment, the nutrient or precursor drug may be absorbed and modified by the carrier organism. In one embodiment, the precursor drug is the basic skeletal structure of aspirin. In another embodiment, the nutrient or precursor drag is, for example, tyrosine, the addition of which promotes the production of gene products such as violacein and other bacterial metabolites which have tyrosine as a precursor.

In an embodiment of any of the above methods, wherein the drops containing one or more test cells contains a sphere coated with the test cells.

In one embodiment of any of the above methods, the drops are 5 or more microns in diameter. In another embodiment, the drops are about 30 microns to about 100 microns in diameter.

In one embodiment of any of the above methods, an oil is used to generate the drops. In a further embodiment, a hydrocarbon oil is used to generate the drops. In a further embodiment, a mineral oil is used to generate the drops. In a further embodiment, a fluorinated oil is used to generate the drops. In another embodiment, the drops contain a surfactant.

In an embodiment of any of the above methods, wherein the sphere is made of Dextran.

In an embodiment of any of the above methods, wherein the sphere is made of gelatin. In a further embodiment, the sphere is made of alginate. In a further embodiment, the sphere is made of an alginate derivative. In a further embodiment, the sphere is made ofan alginate mixed with an additional agent. In a further embodiment, the sphere is made of, the additional agent is a cell binding peptide. In a further embodiment, the additional agent is a magnetic material.

In an embodiment of any of the above methods, of any of the above methods the metagenomic DNA containing carrier organism is administered ballistically. In a further embodiment, the metagenomic DNA containing carrier organism is administered ballistically with a helium powered gene gun.

In one embodiment of any of the above methods, the organism is capable of invading a second organism. In one embodiment of any of the above methods, the organism contains a bacterial invasion gene. In another embodiment, the organism contains a bacterial invasion gene cluster. In another embodiment, the organism contains a bacterial invasion gene operon. In a further embodiment, the bacterial invasion gene is a Salmonella gene. In a further embodiment, the bacterial invasion gene is a Y. pseudotuberculosis gene. In a further embodiment, the bacterial invasion gene is a M. tuberculosis gene. In a further embodiment, the organism expresses viral cell proteins which allow for invasion. In a further embodiment, the organism expresses mammalian cell proteins which allow for invasion.

In one embodiment of any of the above methods, the metagenomic DNA is 5kbp or more in length.

In one embodiment of any of the above methods, the organism is a bacteria. In a further embodiment, the bacteria is an attenuated E. coli. In a further embodiment, the bacteria is of the genus Bacillus. In a further embodiment, the bacteria is of the genus Streptomyces. In a further embodiment, the bacteria is of the genus Actinomyces. In a further embodiment, the bacteria is Agrobacterium tumefaciens. In a further embodiment, the bacteria is Burkholderia graminis. In a further embodiment, the bacteria is Caulobacter vibrioides. In a further embodiment, the bacteria is Pseudomonas putids. In a further embodiment, the bacteria is Ralstonia metallidurans.

In one embodiment of any of the above methods, the test subject is a nematode. In another embodiment, the test subject is a mammalian cell. In another embodiment, the test subject is an amoeba. In another embodiment, the test subject is a mouse. In another embodiment, the test subject is a mouse cell. In another embodiment, the test subject is a human cell. In another embodiment, the test subject is derived from a mammalian cell. In another embodiment, the test subject is capable of ingesting the carrier organism. In another embodiment, the test subject is capable of being infected by the carrier organism. In another embodiment, the test subject is a unicellular eukaryote. In another embodiment, the test subject is a multicullular eukaryote.

In one embodiment of any of the above methods, the organism is a fungal cell. In one embodiment of any of the above methods, the organism is a parasitic protozoa.

In one embodiment of any of the above methods, the expression of the metagenomic DNA results in the organism exhibiting a desired phenotype in a cell from an organism other than the source of the heterologous metagenomic DNA.

In one embodiment of any of the above methods, when the organism is in a nematode the expression of the heterologous metagenomic DNA results in the nematode exhibiting a desired phenotype. In another embodiment, when the organism is in a mammalian cell the expression of the heterologous metagenomic DNA results in the mammalian cell exhibiting a desired phenotype. In another embodiment, when the organism is in an amoeba the expression of the heterologous metagenomic DNA results in the amoeba exhibiting a desired phenotype. In another embodiment, when the organism is in a mouse the expression of the heterologous metagenomic DNA results in the mouse exhibiting a desired phenotype. In another embodiment, when the organism is in a mouse cell the expression of the heterologous metagenomic DNA results in the mouse cell exhibiting a desired phenotype. In another embodiment, when the organism is in a mammalian cell the expression of the heterologous metagenomic DNA results in the mammalian exhibiting a desired phenotype. In another embodiment, when the organism is in a human cell the expression of the heterologous metagenomic DNA results in the human cell exhibiting a desired phenotype. In another embodiment, when the organism is in a cell that is capable of ingesting the organism the expression of the heterologous metagenomic DNA results in the cell that is capable of ingesting the organism exhibiting a desired phenotype. In another embodiment, when the organism is in a cell that is capable of being infected by the organism the expression of the heterologous metagenomic DNA results in the cell that is capable of being infected by the organism exhibiting a desired phenotype. In another embodiment, when the organism is in a unicellular eukaryote the expression of the heterologous metagenomic DNA results in the unicellular eukaryote exhibiting a desired phenotype. In another embodiment, when the organism is in a muulticellular eukaryote the expression of the heterologous metagenomic DNA results in the multicellular eukaryote exhibiting a desired phenotype. In one embodiment of any of the above methods, the desired phenotype can be assessed in an assay. In another embodiment, the desired phenotype can be assessed by use of a stain. In another embodiment, the desired phenotype can be assessed by the activation of a reporter gene. In another embodiment, the desired phenotype can be assessed in an activity assay. In another embodiment, the desired phenotype is cell survival. In another embodiment, the desired phenotype is differentiation. In another embodiment, the desired phenotype is dedifferentiation. In another embodiment, the desired phenotype is cell migration. In another embodiment, the desired phenotype is DNA repair. In another embodiment, the desired phenotype is apoptosis or cell death. In another embodiment, the desired phenotype is cell growth. In another embodiment, the desired phenotype is cell multiplication. In another embodiment, the desired phenotype is cell multiplication, such as increasing the number of liver cells in a dish instead of obtaining stem cells.

In one embodiment of any of the above methods, the desired phenotype is the expression of a reporter gene. In a further embodiment, the reporter gene produces a colored product. In a further embodiment, the reporter gene encodes red, green or yellow fluorescent protein.

In one embodiment of any of the above methods, the heterologous metagenomic DNA is present in a soil sample. In another embodiment, the heterologous metagenomic DNA is present in a water sample. In another embodiment, the heterologous metagenomic DNA is present in a library of DNA from mixed sources. In another embodiment, the heterologous metagenomic DNA is present on the surface of a plant. In another embodiment, the heterologous metagenomic DNA is present in a microbial community associated with an organism.. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that is present on the surface of human skin. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that is present in the alimentary tract of a mosquito. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that is present in the alimentary tract of a mouse. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that is present in the alimentary tract of a caterpillar. In a further embodiment, the heterologous metagenomic DNA is present in a microbial community that has a symbiotic relationship with an organism, such as a bacteria that produces compounds that are toxic in high doses to provide defense to its host, such as a marine sponge. to one embodiment of any of the above methods, the heterologous metagenomic DNA is between 500bp and 250kbp in length.

In an embodiment of any of the above methods, the carrier organism contains heterologous metagenomic DNA encoding a single metabolic pathway. In another embodiment, the single metabolic pathway produces a known drug. In a further embodiment, the DNA encoding a single metabolic pathway is prone to mutation. In another embodiment, the carrier organism is treated with a mutagen.

In one embodiment of any of the above methods, the organism has been modified to allow endosomal escape via coating with recombinant Listeriolysin O.

It is an object of the current invention to provide tools and methods for screening for compounds with biological activity in eukaryotic cells which are normally produced by bacteria found in an ecological niche like a soil sample without being limited by the fact that the vast majority of bacteria cannot be cultured using known laboratory methods and formulas. It is also an object of the current invention to provide methods with sufficiently high throughput so that the enormous genetic variety and the resulting myriad compounds which can come from a single sample can be screened in a reasonable amount of time. It is a further object of the current invention to provide methods which employ microfluidics and also methods of comparable efficiency which do not.

These methods involve the screening of metagenomic libraries and do not require culturing the bacteria found in the primary sample or the generation of extracts of bacterial metabolites. Metagenomic libraries are created by isolating high molecular weight DNA from the environmental sample (e.g. soil) and then reducing the DNA to a predetermined size range (e.g. 15-30 kb) through the partial digestion of the high molecular weight with a restriction enzyme that leaves overhangs. The fragments of digested metagenomic DNA are then ligated into an expression vector which typically has artificial constitutive (e.g. T5) or inducible (e.g. lac) promoters flanking both ends of the insert. The promoters drive expression in both directions to ensure expression of any bacterial operons that lack promoter activity. The metagenomic libraries typically contain at least ten million and up to more than one hundred million elements derived from all of the bacterial DNA in a sample. The inserts in any metagenomic library are usually of a narrow range of sizes with the average size depending on the method used to digest the bacterial DNA. The minimum size is usually set at about 15 kb to insure that there is a reasonable probability of having the minimal elements necessary to produce a single metabolite on one operon. In practice, the maximum size of the inserts has been about 250 kb when BACmids were used.

In the current invention, metagenomic libraries are used to transform common, readily culturable laboratory bacteria like E. coli so that each cell in a sample contains at most about one insert from the metagenomic library. The bacteria used in the current invention are capable of invading eukaryotic cells which includes mammalian cells, particularly human cells and especially human cancer cells (Fig. 1). They also include, without limitation, fungal cells like those of the Candida genus and parasitic protozoa like those of the Plasmodium genus. Invasive bacteria, in particular coli, occur naturally (Carvalho, 2009) or they can be made invasive by inserting a bacterial invasion gene such as invasin from Y. pseudotuberculosis or Mcela from M. tuberculosis, both of which can be expressed on the surface of the E. coli.

Elements of the current invention are cultures of invasive bacteria, particularly E. coli, which have been transformed with a metagenomic library so that each organism in the culture contains at most about one insert from the metagenomic library. Another element of the current invention are eukaryotic cells which may be engineered to contain a reporter gene which is expressed and/or activated when a metabolite produced from the bacteria leads to a desired physiological change. For example, the eukaryotic cell could be a human cancer cell engineered to express red or green fluorescent protein when a bacterial metabolite induces apoptosis in the cell. In the microfluidic based method of the current invention, the two key elements, the invasive bacteria and the engineered eukaryotic cell are brought together in a fashion which is designed to insure that each eukaryotic cell in invaded by multiple copies/clones of the same bacteria that carry the same fragment of metagenomic DNA (Fig. 2). In the direct exposure embodiment of the current invention, invasive bacteria are mixed with mammalian cells at a ratio of one bacteria per mammalian cells. Upon infection, the invasive bacteria (either naturally or through engineering) will multiply as a clonal population within the mammalian target cell - essentially converting each mammalian cell into a microwell.

Using microfluidics, this is a straightforward process involving the generation of two streams of microdroplets (Fig. 3). In the first stream, each droplet is formed to contain only one bacterium which may subsequently divide but the result is that all of the bacteria in any given droplet are genetically identical. In the second stream, each droplet is formed around a single engineered eukaryotic cell. The first stream and the second stream are brought together and one microdroplet of the first stream is fused with one microdroplet of the second stream. Droplet merger channels currently in use generate about 30,000 drops an hour and it is not uncommon to have ten droplet merger channels on a single chip. Each of these droplets is equivalent to a well in a microtiter plate so that one chip generates 300,000 wells or experiments per hour. By controlling the relative concentration of bacteria and eukaryotic cells and limiting the time they are exposed to each other, each eukaryotic cell is invaded by a determined average number of bacteria. The exposure is terminated by killing all of the extracellular bacteria which have not invaded the eukaryotic cells by adding gentamycin or another antibacterial which does not penetrate the eukaryotic cells.

In the preferred embodiment of the invention eukaryotic cells are exposed directly to invasive bacteria that have not been grown in microfluidic droplets. The bacteria are instead mixed at a ratio of approximately 1-2 bacterium per eukaryotic cell for the infection process. Bacteria will have been engineered so that upon being internalized by the eukaryotic cell the bacteria will multiply within each mammalian cell. This has been demonstrated with adherent-invasive E. coli (AIEC) which have been shown to invade and multiply within the cancer cell line J774A.1. As the internalized bacteria multiply they will produce metabolites encoded by the metagenomic DNA. The metabolites will either be exported into the cytoplasm of the eukaryotic cell by the bacteria or in certain cases eukaryotic cell permeable bactericidal agents such as ciprofloxacin will be added to the culture media to induce lysis on a portion of the intracellular bacteria, allowing for the release of their novel metabolites.

Each engineered eukaryotic cell whether formed by the microfluidic method or the direct exposure method contains a population of bacteria which are genetically identical. These cells are an additional key element of the invention and each is equivalent to a microtiter well and each contains a single screening experiment. The infected eukaryotic cells are subsequently cultured for a period ranging from 4-72 hours and then passed through a fluorescence activated cell sorter (FACS) which separates and isolates eukaryotic cells in which the reporter gene has been activated. Alternatively, eukaryotic cells undergoing a physiological change and displaying a cell surface marker(s) may be isolated through the use of magnetic beads conjugated to antibodies or antigens specific for those eukaryotic cell surface markers. The isolated eukaryotic cells are then lysed as a batch or as individual separated cells and the internalized, drug producing bacteria plated on agar plates or inoculated in liquid media, cultured, and the metabolites produced collected and analyzed first by mass spectrometry. By comparing the metabolite mass spectrum from the metabolite producing bacteria with the control bacteria, it is possible to identify the molecular ion peaks resulting from the DNA insert in the bacterium. The insert can also be sequenced to provide information about the enzymes present which sheds light on the structure of the new metabolites. Fragmentation mass spectrometry and NMR can be used to complete the elucidation of their structures.

The following Examples are set forth to aid in an understanding of the subject matter of this disclosure, but are not intended to, and should not be construed to, limit in any way the claims which follow thereafter.

EXAMPLE 1

Initial experiments involved the delivery of violacein drug via E. coli expressing the violacein operon (VioA-E) from Chromobacterium violaceum (Fig. 4A) into cervical cancer cell line HeLa S3. The 8 kbp violacein operon was cloned into a pUC19 high copy plasmid and transformed into DH10B E. coli that already contained a secondary plasmid expressing invasin. As seen in Figure 4 bacterial delivery of violacein expressing into HeLa S3 cells results in widespread cell death, unlike treatment with non-invasive violacein expressing bacteria (Fig. 4B). This death confirmed to be apoptosis (Fig. 5).

The results demonstrated were accomplished with a standard lab strain of E. coli used for general plasmid cloning and may be coupled for microfluidics for high throughput screening. For the direct exposure method there are several options, though the invention is not limited to these reagents. One method is to use an adherent-invasive E. coli (AIFC) type strain such as LF82 which has been demonstrated to replicate in the inacrophage-like cancer cell line J774A.1. By adding invasin to these bacteria we were able to achieve a more consistent and widespread pattern of infection and replication in J774A.1 cells (Fig. 6).

A second method involves engineering E. coli with the M. tuberculosis protein Mcela, which has been demonstrated to impart an invasive ability to E. coli as well as allow for intracellular replication of the £. coli. A third method involves expressing the Y. pseudotuberculosis protein invasin as well as the S. flexneri protein IcsA in iE. coli and finally coating the E. coli in recombinant listeriolysin O (LLO) to achieve invasion, endosomal escape and intracellular replication.

Protocols for the invention

Microfluidics method:

1. Isolate metagenomic DNA or acquire ready-made metagenomic plasmid DNA library containing fragments ~5 kbp or greater.

2. Transform metagenomic library into bacterial strain of interest (e.g. nonpathogenic E. coli) which has been modified to express a bacterial invasion gene (inv) from Y. pseudotuberculosis or another source microbe. The bacteria may express the protein from a plasmid or as a genomic DNA insertion.

3. Grow invasive bacteria carrying metagenomic library to desired density.

4. Wash invasive bacteria carrying metagenomic library and dilute in serum free media.

5. Pump invasive bacteria carrying metagenomic library along with a parallel stream of biocompatible oil into a microfluidic chip with droplet forming channels.

6. Form droplets with an average of one bacteria per drop.

7. Grow bacteria in droplets through simply placing droplets at 25-37°C with or without light agitation until desired bacterial density is reached.

8. Pump droplets containing clonal populations of bacteria into a microfluidic chip in which mammalian cells are concurrently formed in serum free media with an average of between 10-100 mammalian cells per drop. Allow pre-formed bacterial cell drops and newly formed mammalian cell drops to migrate through their respective channels which are designed to merge. By choosing the appropriate flow rate for each channel bacteria and mammalian cells will be arranged to enter the merger channel in a 1 : 1 ratio. As the drops proceed further along the channel they will enter a merger element that fuses pairs of bacterial and mammalian cells drops via mechanical forces (Fig. 3) or through electrofusion (not shown).

Merged droplets will be incubated at 37°C to allow bacterial invasion of target mammalian cells.

Post- infection merged bacterial-mammalian droplets will be disrupted with a combination of dilute detergent and mild agitation. Mammalian cells will be washed to remove any trace of detergent and resuspended in culture media supplemented with a non-mammalian cell permeable antibiotic such as gentamicin to kill off remaining exogenous bacteria in the culture media.

Mammalian cells, each carrying a unique clonal population of bacteria expressing metagenomic fragments of DNA, would then be incubated between 32-42°C between 4-72 hours to allow intracellular bacterial lysis, release of unique metabolites produced from metagenomic DNAs, and phenotypic effect on the target mammalian cells.

Isolate mammalian cells through antibody/antigen binding and secondary fluorescent labeling or affinity labeled magnetic beads. Alternatively, engineer mammalian cells to express a reporter gene such as red or green fluorescent protein that is produced during a particular physiological state. Alternatively, add a cell permeable reporter molecule or peptide (e.g. caspase activated FLICA from Invitrogen) that is modified and activated when cells undergo a particular physiological change such as undergoing apoptosis.

Upon isolating affected mammalian cell population of interest, lyse mammalian cells in a solution of 0.1-1% triton X-100 to release remaining, intact intracellular bacteria.

Pick individual bacterial colonies and retest the functionality of each in separate wells of multiwell plates.

Isolate metagenomic plasmids from invasive bacteria that are successful for their second round of tests. Perform several diagnostic restriction digests to determine which clones arc unique. Further analyze unique clones through sequence analysis.

17. Bacterial strains carrying unique metagenomic plasmid DNAs that have produced effective metabolites can be further analyzed by Maldi-TOF to versus control strains to determine the unique metabolites produced. Combined with plasmid mutagenesis specific metabolites responsible for induced phenotype can be determined.

Direct exposure method:

1. Isolate metagenomic DNA or acquire ready-made metagenomic plasmid DNA library containing fragments ~5 kbp or greater.

2. Transform metagenomic library into intracellular multiplying bacterial strain of interest. The strain can either be naturally occurring (e.g. adherent-invasive E. coli) or a modified strain (e.g. BL21 E. coli expressing Y. pseudotuberculosis invasin and S. flexneri IcsA and further coated with recombinant listeriolysin O (LLO). Modified strains may express the protein(s) from a plasmid or as a genomic DNA insertion.

3. Grow invasive, intracellular replicating bacteria carrying metagenomic library to desired density.

4. Wash invasive, intracellular replicating bacteria carrying metagenomic library and dilute in serum free media.

5. Mix invasive, intracellular replicating bacteria and target mammalian cell line at a 1:1 to 1:10 ratio so that each mammalian cell receives, on average 1-2 invasive bacteria per cell.

6. Allow invasion to occur for 10 minutes-2 hours at 25-37°C.

7. Post-invasion wash cells in media containing gentamicin or a similar non- mammalian cell permeable antibiotic to kill off remaining exogenous bacteria in the culture media.

8. Incubate mammalian cells at 32-42°C and allow intracellular bacteria to multiply. If intracellular bacteria reach a certain number (e.g. 30 bacteria per mammalian cell) and replication needs to be halted a mammalian cell-permeable bacteriostatic agent such as tetracycline may be added). 9. Allow incubation to continue until bacteria are lysed or pulse mammalian cells with a bactericidal agent such as ciprofloxacin. Upon lysis novel metabolites produced from metagenomic DNAs will be released into the host mammalian cell.

10. Isolate mammalian cells through antibody/antigen binding and secondary fluorescent labeling or affinity labeled magnetic beads. Alternatively, engineer mammalian cells to express a reporter gene such as red or green fluorescent protein that is produced during a particular physiological state. Alternatively, add a cell permeable reporter molecule or peptide (e.g. caspase activated FLICA from Invitrogen) that is modified and activated when cells undergo a particular physiological change such as undergoing apoptosis.

11. Upon isolating affected mammalian cell population of interest, lyse mammalian cells in a solution of 0.1-1% triton X- 100 to release remaining, intact intracellular bacteria.

12. Pick individual bacterial colonies and retest the functionality of each in separate wells of multiwell plates.

13. Isolate metagenomic plasmids from invasive bacteria that are successful for their second round of tests. Perform several diagnostic restriction digests to determine which clones are unique. Further analyze unique clones through sequence analysis.

14. Bacterial strains carrying unique metagenomic plasmid DNAs that have produced effective metabolites can be further analyzed by Maldi-TOF to versus control strains to determine the unique metabolites produced. Combined with plasmid mutagenesis specific metabolites responsible for induced phenotype can be determined.

Example 2

Introduction

General

Over the last two decades there has been a gradual decrease in the number of new chemical entities (NCEs), that is, drugs with completely novel chemistries. Though significant advances have been made with combinatorial chemical synthesis, these are often inferior to compounds found in nature which have evolved to bind protein motifs that are conserved throughout all animal kingdoms. A potentially rich source of natural compounds is found within complex communities of microbial organisms whose DNA is often referred to as the metagenome. Metagenomic DNA is comprised of the genomic DNA of many microorganisms in an environment, many of which cannot be cultured using standard methods. By cloning large pieces of genomic DNA from these microorganisms it is possible to capture entire regions coding for enzymes that produce novel compounds.

A major limitation in the current methodologies used for mammalian screening of metagenomic libraries is that extracts must be prepared and tested in a multiwell format. This procedure is laborious and greatly limits the number of clones and compounds that can be screened. Here we develop a bacterial-based delivery method that bypasses the need for extract preparation in order to deliver metagenomic derived drugs into mammalian cells. Combined with microfluidic technology, our methodology reduces the costs associated with standard screening practices, eliminates the need for mutliwell plates, and identifies natural products that produce desired in vitro and/or in vivo effects on an unprecedented scale.

Microbes as a Source of Natural Products

The identification of natural compounds produced by microbes isolated from various environments has persisted for decades and has led to such success stories as rapamycin and FK-506 (tacrolimus) (Kino et al. 1987; Paghdal et al. 2007; Vezina et al. 1975). However, the majority of microbes, greater than 99%, remain unculturable (Lewis 2007; Saleh-Lakha et al. 2005; Torsvik et al. 1990). To gain access to the unculturable majority, a variety of approaches have been taken. In the most straightforward approach, trial and error can eventually lead to the proper combination of growth media and conditions needed to cultivate a newly identified bacterial strain (Davis et al. 2005; Stevenson et al. 2004). In a more recent technological development it was demonstrated that isolating individual microbes in microdroplets within a gel emulsion allowed for the isolated growth of a variety of bacterial species that otherwise could not be cultured en masse (Zengler et al. 2002). Unfortunately, neither of these methodologies is suitable for the large scale identification of novel compounds in mammalian cell assays. The first method is time consuming and limited in scope while the latter method does not provide a means for easy incorporation into mammalian cell culture assays. Additionally, both methods lessen the chances of identifying enzymatic activities and/or natural compounds that are only transiently produced in response to specific stimuli.

What has become the most prevalent method for mining useful enzymatic activities and compounds from uncharacterized microbes involves bypassing the need for direct culture altogether (Rondon et al. 2000; Treusch et al. 2004). Genomic DNA is extracted from an environmental sample (e.g., soil) and used to generate plasmid libraries which are then propagated and expressed in bacterial strains with simple growth requirements. This method has yielded some success in isolating medicinal natural compounds (Gillespie et al. 2002; Geng et al. 2008) but the majority of screens using such libraries have primarily focused on isolating enzymes with industrial value (Knietsch et al. 2002; Li et al. 2008). Furthermore, there has been very little done to screen such libraries in live mammalian cells. This is primarily due to the fact that the number of positives to come out of such screens is typically very low and the number of individual mammalian cell wells that would have to be screened would be in the hundreds of thousands, if not millions. Added to the challenge is the inability of any one extraction agent to act as an efficient solvent in one step for all novel compounds (Kim et al. 2009), necessitating repeated extractions and screenings for each library.

Microfluidics for High Throughput Mammalian Cell Assays

Microfluidic chip technology has proven to be a very versatile and a reliable means to miniaturize many standard technologies available to researchers (thus earning the name "lab- on-a-ehip" (Chovan et al. 2002; Weigl et al. 2003)). Early emulsion-adapted assays such as in vitro translation and droplet based PCR demonstrated that large scale enzyme based microdroplet assays were feasible and highly reproducible. These laid the foundation for new technologies such as 454's high throughput parallel sequencing (Doi et al. 1999; Margulies et al. 2005).

Recent studies have gone on to demonstrate that conditions within microdroplets can be suitable for the survival and growth of microorganisms such as bacteria (Huebner et al. 2007) and yeast (Luo et al. 2006). The incorporation of gas permeable fluorocarbonated oil as well as biocompatible surfactants has allowed for the survival and growth of mammalian cell types such as 2C6 hybridoma cells and the T cell leukemia cell line, Jurkat (Clausell-Tormos et al. 2008; Koster et al. 2008). Remarkably, similar conditions combined with an increase in droplet size have allowed for the encapsulation and even proliferation of the model organism C. elegans (Clausell-Tormos et al. 2008; Shi et al. 2008). These latest advances would not have been possible without the advent of microfluidic chips. The mixing of oil and cell media at determined flow rates through channels etched into glass or polydimefhylsiloxane (PDMS) chips has made it possible to generate steady streams of microdroplets of a predetermined size and to manipulate these droplets once formed. For example, by forcing droplets along a temperature controlled path it was possible to selectively freeze drops containing single mouse B lymphocytes (Sgro et al. 2007); a procedure that would otherwise have been cumbersome and far less efficient if it weren't for the ability to wholly contain the procedure within a chip. Thus, an almost endless number of existing experimental conditions can be adapted to microfluidic droplet format. Any minor deviations in standard on-the-bench protocols have been more than offset by the fact that researchers utilizing microfluidic chips can perform hundreds of thousands or even millions of experiments an hour.

Experiments have found suitable microfluidic chip designs and reagents that would allow for cell and organism based drug screening. To date, microfluidic chip technology has yet to be used for drug screening in mammalian cells. It has been suggested that traditional high throughput screening (HTS) could be coupled with microdroplet screening if small molecule libraries with diffusion resistance oils were utilized (Courtois et al. 2009; Huebner et al. 2008). This would certainly miniaturize currently available HTS assays but would confine researchers to the limitations inherent to small molecules produced by combinatorial chemistry. Though natural product based mammalian drug screens have been suggested (Koster et al. 2008 and RainDance Inc., unpublished observations) no satisfactory methodology has been offered up that would allow for efficient screening of any of the vast number of chemical compounds found in nature.

Apoptosis of Cancer Cells

Apoptosis is an orderly process of programmed cell-death (PCD) which is characterized by a defined set of events that result in the safe disposal of tens of billions of unwanted cells from our bodies on a daily basis. Successfully executed, apoptosis prohibits unwanted retention of cells which could lead to life threatening diseases such as autoimmune disorders and cancer (Elmore 2007). In the case of cancer, the apoptotic process is derailed during cellular transformation and the result is a population of cells that have become desensitized to normal apoptotic stimuli (Call et al. 2008).

Standard therapeutic regiments attempt to reinitiate apoptosis but in some cases, cancerous cells accrue additional mutations that render them resistant to drug and/or radiation induced apoptosis. The latter phenomenon is known to occur through several mechanisms. These include broad resistance mechanisms such as multidrug resistance (MDR) whereby cellular proteins (e.g., P-glycoprotein) act as efflux pumps to rid cells of drug compounds before they can take effect on their targets (Goda et al. 2009). Alternatively, specific mutations and gene deregulation can also lead to inhibition of therapy induced apoptotic pathways in cancer. These can include the inactivation of proapoptotic factors such as p53 (Xue et al. 2007) or the p53 pathway (Tweddle et al. 2003) as well as induction of antiapoptotic factors such as Bcl-2 (Gazitt et al. 1998) and BCL-X(L) (Ding et al. 2000; Vogler et al. 2009). Resistance to anticancer compounds can even come about by direct mutation of drug target proteins such as the mutations in Bcr Abl tyrosine kinase which render chronic myeloid leukemia cells resistant to induction of apoptosis by Imatmib (Gleevec) (Shah et al. 2002). Knowledge of specific mutations has been exploited in attempts to, for example, restore the function of known proapoptotic factors such as p53 through the use of small molecule drugs (Weinnman et al.

2008) , thus overcoming drug resistance. While this type of strategy has shown some promise the technology has yet to be proven successful in the clinic. On the other hand, there are a couple of recent instances where natural products have been successfully used to overcome drug resistance and to initiate the apoptotic program. In 2007 ixabepilone, a semisynthetic analog of microbially produced epot ilone B, successfully entered the market for the treatment of metastatic breast cancers resistant to the plant-derived anticancer drug taxane (Dumontet et al. 2009). In the same year temsirolimus, a derivative of rapamycin, was approved for treatment of renal cell carcinoma, a cancer that is resistant to conventional chemotherapy (Prenen et al.

2009) . Currently in clinical trials, a derivative of the microbially derived drug geldanamycin, alvespimycin (Kurashina et al. 2009) has already shown promise in treating metastatic breast cancers as well as leukemia's highly refractory to standard treatment. Though potentially subject to drug resistance, the discovery of microbial derived natural products would add new drugs and chemistries to the arsenal of anticancer compounds that could target inhibitors of apoptotic pathways.

Microbes produce a wide range of natural products that have proven valuable for the therapeutic purposes. The ability to produce metagenomic libraries in principle provide an opportunity to produce and screen large numbers of natural products, although methods to do so are lacking. We have devised a method for the efficient transfer of large numbers of metagenomic clones into mammalian cells. When coupled with microfluidics and our expertise in high throughput methods, we believe these technologies will form a powerful approach for screening large numbers of natural products for pro-apoptotic anticancer activities.

We have extensive experience in development of high throughput methods and have recently developed a novel approach for the efficient transfer and screening of natural products in mammalian cells. Experiment

Establishment of an Extraction-Free Method for Mammalian Cell Delivery of Microbtally Produced Natural Product

We have successfully developed a method for natural product delivery into mammalian cells that functions regardless of a compound's extraction and solubility requirements; two major hurdles for high-throughput mammalian cell screening. As noted earlier one sticking point in the search for microbe produced natural products is the need to use various extraction methods, none of which can act as a universal solubilizmg agent. We devised a simple method which allows for direct delivery of the bacteria producing a natural product into a mammalian cell (Fig. 7A). To do this, we generated a i as mid (TRIP 3.0) that would enable a standard laboratory strain of E. coli to invade mammalian cells. As previously reported (Grillot-Courvalin et al. 1998; Isberg et al. 1987), the Y. pseudotuberculosis protein invasin confers on E. coli the ability to invade both dividing and quiescent mammalian cells. The expression of invasin in TRIP 3.0 (Fig. 7B) is driven by a constitutive promoter. Since it has been repeatedly demonstrated that the L. monocytogenes listeriolysin O (LLO) protein allows for the most effective delivery of bacterial payload (e.g., plasmid), we incorporated an additional ribosome binding site (RBS) and the LLO gene immediately downstream of invasin. The presence of the RBS allowed for the elimination of a large amount of upstream and downstream sequence that was retained in a previously published plasmid dependent on multiple promoters (Xiang et al. 2006). Lastly, we incorporated a mid-copy origin of replication, pACYC, to avoid recombination between TRIP 3.0 and the origins of any secondary plasmids shuttled into our E. coli strain of choice. Successful invasion was determined through the use of a gentamicin protection assay (Hess et al. 2003), which uses gentamicin to eliminate non- internalized bacteria. E. coli strain DH10B (Invitrogen) expressing TRIP 3.0 was found to efficiently invade two different mammalian cell lines - Vero African green monkey kidney epithelial cells and HeLa S3 human cervical carcinoma cells.

In order to test the ability of invasive bacteria to deliver natural products we chose to work with a fragment of the C. violaceum genome encompassing the violacein operon. Violacein is a pigmented antibiotic which gives C. violaceum colonies a dark violet color. It has been determined that violacein has evolved as a defense for C. violaceum biofilms and that phagocytosis of a single C. violaceum will result in lysis, release of violacein, and induction of apoptosis in protozoa (Matz et al. 2008). The violacein operon is comprised of the genes VioA-E, the products of which are sufficient for the complete synthesis of vio!acein in the presence of tryptophan and oxygen substrates. Additionally, violacein has been shown to be toxic to not only protozoa but a wide range of mammalian cell lines. For this work, the violacein operon was cloned into an Ampicillin resistant vector with a high copy origin of replication and subsequently transformed into DH10B cells already carrying plasmid TRIP 3.0. Violacein production was evidenced by the dark violet colonies of single and dual transformants grown on selective media (Fig. 8A). Double transformants were tested in parallel with appropriate controls for their ability to exert an effect on mammalian cell lines. Overnight bacterial cultures were diluted and spun onto monolayers of near confluent cells to allow for infection. It was observed that even at a low multiplicity of infection (MOI) invasive bacteria expressing violacein resulted in detachment of the majority of HeLa S3 cells within 20 hours (Fig. 8B, lower right). Similar results were obtained for Vero cells (not shown). Control bacteria expressing invasin demonstrated mild toxicity at an MOI of 75, well above the effective infection levels.

The timing of the cell-death following internalization of violacein producing bacteria was more thoroughly examined in HeLa S3 cells. MOIs of 10 and 25 produced the widespread appearance of vacuoles in HeLa S3 cells at 4 hours post-infection. As shown in Figure 9, only cells treated with invasive bacteria producing violacein developed the presence of refractive, vacuole-like structures. At 6 hours post-infection live cells were stained with Armexin V-fluorescein isothiocyanate (Annexin V-FfTC Apoptosis Kit, CalBiochem). A low percentage of cells at MOI 10 (not shown) stained positively for Annexin V and even more stained positive at MOI 25 (Fig. 9). Similar to empty vector and uninvasive violacein expressing bacteria, HeLa S3 cells treated with control invasive E. coli were clearly negative for anything beyond background staining (Fig. 9, bottom row, left panel). We were able to reproduce the appearance of Annexin V- FITC positive cells with purified violacein (Sigma) when it was added to HeLa S3 cells at a concentration range of 1-3 DM (not shown). The differences we noted with purified violacein were that the appearance of multiple large vacuoles did not occur and Annexin V-FITC staining took several hours longer to appear, was more widespread but less intense. This could possibly be due to variability in the exact number of violacein producing bacteria to invade HeLa S3 cells.

Lastly, we tested the feasibility of recovering violacein producing invasive bacteria from plates of cells undergoing apoptosis or having undergone violacein induced apoptosis. Initially, we compared the rescue of violacein producing bacteria versus controls in wells of treated HeLa S3 cells observed to be undergoing apoptosis via Annexin-V staining. Even at 10 hours post-invasion it was possible to rescue comparable numbers of violacein producing bacteria from HeLa S3s undergoing morphologic changes and staining for positive for Annex in V-FITC as it was to rescue invasive bacteria controls (data not shown). These results were followed with a late stage rescue of violacein producing bacteria from HeLa S3 cells that were detergent lysed 24 hours post bacterial exposure at a MOI of 10. We specifically compared two populations - cells treated with invasive and noninvasive violacein producing bacteria. Approximately six fold (-600 total colonies) more bacteria were recovered from infected HeLa S3 than from a confluent layer of control cells treated with noninvasive violacein producing E. coli. Identical to bacteria recovered from populations of HeLa S3 cells undergoing early and mid-stage apoptosis, 100% of late-stage recovered bacteria expressed violacein as determined by the dark violet pigmentation of colonies. Though there was some level of background, the number of HeLa S3 cells in control wells were well above the violacein treated wells (Figure 8B, compare upper left and lower right panels).

Optimization of Microdroplet Formation

To successfully execute high throughput screening with natural product producing invasive bacteria it was necessary to establish a method of compartmentalizing individual bacteria. In collaboration with Dolomite Inc. we conducted experiments aimed at choosing the optimal surfactants, chip design and droplet forming conditions for bacteria and mammalian cells. For bacterial droplet formation a Dolomite Mitos Syringe Pump XS was used to deliver media into a T-junction chip (Fig. 10) to determine optimal conditions for stable drop formation. We found that with the fluorocarbonated oil FC-40 and bacterial growth media Luria Broth we could form stable -1-2 nL sized droplets at various flow rates (Table 1) and that we could easily achieve production of over 500 droplets per minute. Similar experiments were carried out with mammalian cell culture media but droplet formation did not occur in the absence of surfactant. The presence of 1% biocompatible surfactant Span 80 allowed for the formation of droplets with regular stability (Fig. 11). We have since determined that the hydrophobic coating was not sufficient for proper droplet formation and that coating surfaces with Aquapel hydrophobic polymer has shown promising initial results.

Table 1. Generation of stable Nanoliter droplets with bacterial growth media. FC-40 oil and Luria Broth were introduced into a glass T-junction chip at various flow rates. Both droplet size and the number of droplets per minute could be controlled by adjusting rates of flow. The results did not vary with the presence of surfactant.

Our studies demonstrate that we can deliver natural products into mammalian cells without the need to purify compounds or even prepare crude extracts. Intact bacteria that continue to produce natural product can easily be rescued from infected mammalian cells for analysis. We further demonstrated the feasibility of encapsulating bacteria in high throughput through the use of microfluidic chip droplet formation.

Materials and Methods

Bacterial culture and plasmids

All E. coli strains were grown in low salt Luria-Bertani Broth (LB). Final concentrations of antibiotics used were ampicillin (100 Og/rnL) and kanamycin (35 Og mL). Vector pET24B containing the violacein operon (VioA-E) was a gift of Jonathan Clardy (Harvard University). The violacein operon was removed by digestion with Mlul and Xhol and shuttled into an ampicillin resistant plasmid with pUC19 origin and type II restriction sites that left Mlul and Xhol overhangs when digested with Bbsl (plasmid designed and provided by DNA 2.0 Inc.) to generate pUC-Vio. Control plasmid (pUC-Re) lacking the violacein pathway was generated by digesting with Bbsl and then blunting ends with Mung Bean Nuclease prior to self-ligation. TRIP 3.0 p!asmid was custom engineered by DNA 2.0 Inc. and is a mid-copy vector with pACYC origin of replication and is kanamycm resistant. TRIP 3.0 expresses the inva in gene and promoter region amplified from Y. pseudotuberculosis genomic DNA. EPI100 cells carrying either pL ^' C-Rc or pUC-Vio were transformed with TRIP 3.0 selected for presence of both plasmids.

Metagenomic Library Preparation

Metagenomic libraries were generated in cos mid vector pWEB by C. Saski at Clemson University Genomics Institute (CUGI). Briefly, genomic DNA was extracted, purified, mechanically sheared to an average size of 40 kbp, ligated into phage adapter ends, packaged in vitro and subsequently used to infect E. colt strain EP1100.

Cell culture

HeLa S3 cells and mouse P19 embryonic carcinoma cells were cultured using standard aseptic technique and were grown in an incubator with 5% C02 and a relative humidity of 60-80%. Media was Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% Heat Inactivated Fetal Bovine Serum and IX Glutamax (Invitrogen). Routine subculture was performed via washing semi-confluent cells with PBS and then dissociating cells with 0.25% trypsin (Invitrogen). For experiments involving cell encapsulation, Turbo DNase (Ambion) was added at a concentration of 2 units/mL. F68 pluronic (Invitrogen) was added to a final concentration of 1%.

For incubations involving bacterial infection of HeLa S3 cells, 100 Dg/mL of gentamicin was used for initial washes and the first hour of incubation. Subsequently, the media concentration of gentamicin was lowered to 20 Dg mL. For recovery of bacteria from sorted HeLa S3 cells, CFU buffer (0.1% Triton-X 100) was used to lyse HeLa S3 for a period of 10 minutes at room temperature.

Microfluidics chips and instrumentation

Polydimethylsiloxane (PDMS) chip designs were generated via Autocad software (Supplement MX= 1). Chips were manufactured through the services of the State of Utah Center of Excellence for Biomedical Microfluidics. Prior to use, chip channels were flushed with a solution of 1% (Heptadecafluoro-l,l,2,2-Tetrahydrodecyl) Trichlorosilane (TSC) v/v in fluorinated oil to generate a hydrophobic coating and prevent wetting of drops. All connections to drops for injection of liquids were done with nonreactive PEEK and/or PTFE tubing. Merging module was completed by injection of molten Low Melting Temperature Indium Alloy #1 17 (McMaster-Carr) into preformed channels in the PDMS chips with syringes fitted with blunt tip needles. Fluorinated oil and media were fed into PDMS chips via a pressure driven pump (Fluigent, France). Electric field for drop fusion was generated with a 14V, 400W power supply. For fusion, a constant voltage setting of approximately 400 V was used. For homogenous suspension of mammalian cells while injecting into PDMS chips, a cell stirrer was used (Cetoni, Germany).

For destabilization of chip generated emulsions, an equal volume of 1 H.I H,2H.2H- Perfluorooctanoal (98%, Alfa Aesar) was used.

Protocol

Dayl

1. Dilute overnight bacterial cultures (ΕΡΠ00 + T IP3.0 + pUC- e and EPI100 + TRIP3.0 + pUC-Vio) to approximately 1-2 bacteria per 0.5 nL

2. Inject bacteria into microfluidics chip bacterial drop generator circuit via pressure driven displacement. Set pressure driven pump to simultaneously inject mixture of fluorinated oil + surfactant to generate drops encapsulating 1-2 bacteria.

3. Collect bacteria-containing drops in glass vials with a layer of oil pretreated with 1% TSC in fluorinated oil.

4. Incubate bacteria drops overnight at 30-37°C, making sure to include a secondary 'bladder' tube to collect any overflow of the lower oil layer during expansion of the fluorinated oil due to temperature shift.

Dav2

1. Wash a near-confluent plate of HeLa S3 cells IX with PBS, trypsinize and count. Pellet cells and then suspend at a concentration of 20 million cells/mL in cell culture media + 1% pluronic F-68 + 2 units/mL Turbo DNase. Run cell suspension through a 40 um cell strainer and then place on ice until cell encapsulation.

2. With a 1 mL luer lock syringe, inject HeLa cell suspension to cell stirrer prior to activating pressure driven pump. Connect PEEK tubing from cell stirrer to PDMS chip at appropriate port in the mammalian and bacterial drop fusion element. Attach glass vial containing encapsulated bacteria from Day 1 (bacteria should now have saturated the drops) to PDMS chip. Begin reinjection of drops containing bacteria (either control or bacteria producing violacein). Immediately after reinjection has begun, begin injection of A) HeLa S3 cells in media with pluronic B) stream of fluorinated oil + surfactant for fresh encapsulation of HeLa S3 cells.

Adjust pressure of channels driving flow of reinjecting bacterial drops, oil + surfactant for bacterial drops, mammalian cells in media and oil + surfactant for drops containing mammalian cells until a proper pairing of drop + bacteria and drop + HeLa S3 cells is achieved. Ideally each individual drop containing HeLa S3 cells should only be paired with a single, smaller drop containing bacteria. Once proper pairing of drops containing bacteria and mammalian cells is achieved the electrical leads can be attached to the exposed wires embedded in the alloy surrounding the merging module. The power source can then be turned on so that an electrical field is applied in order to begin merging of each pair of drops that enters the merging module.

Continue to fuse paired drops and collect in an Eppendorf tube that has a small layer (e.g. 50-100 DL) of fluorinated oil + surfactant in order to properly view the collection of drops into the tube. Once a sufficient number of merged drops have been collected (e.g. one million merged drops) incubate the drops at 37°C in a humidified C02 incubator to allow for infection of the HeLa cells by the EPI100 cells expressing invasin (TRIP 3.0) and control (pUC-Re) or violacein (pUC-VIO) plasmids.

After 30-60 minutes of infection time, break open the drops/emulsion to release infected HeLa S3 cells as well as non-internalized bacteria with an equal volume 1H,1H,2H,2H-Perfluorooctanoal. Once the emulsion has been destabilized and formed two layers (an upper layer with media and cells and a bottom layer of oil and surfactant) remove the upper layer and transfer to an Eppendorf tube.

Wash HeLa S3 cells several times with cell culture media + 100 Dg/mL getnamicin in a microfuge set at approximately 2000 rpm to efficiently pellet infected mammalian cells while leaving a majority of the non-internalized bacteria in the supernantant. Once HeLa S3 cells have been washed, transfer them in media + 100 Dg mL gentamicin to a 37°C C02 incubator and leave undisturbed for one hour. Following the one hour incubation, pellet HeLa S3 cells and resuspend them in new media containing 20 Dg mL gentamicin. Incubate cells for the duration of the assay at 37°C in humidified C02 incubator. 9, When the cells have been incubated for a sufficient amount of time, harvest the cells via trypsinization. Stain cells with fluorescent marker of choice (e.g. for apoptosis, fluorocrome inhibitors of caspase activation (FLICA)). At this point, in the control experiment, HeLa S3 cells that have been treated with invasive EPI100 cells producing violacein can be 'spiked' into HeLa S3 cells treated with control EPI100 cells that are not producing drug. Finally, sort cells with the highest fluorescence intensity via fluorescence activated cell sorting (FACS).

10. Harvest cells that are identified as positives and lyse immediately in CFU buffer. Finally, plate lysate on LB -agar plates containing the appropriate antibiotics (in this case, minimally containing ampicillin). Determine the efficiency of the retrieval of HeLa S3 cells having been infected with EPIIOO cells producing violacein.

Example 3

Our method involves a way to screen metagenomic libraries without the need for generating bacterial cell extracts. The method also involves linking the novel drug to the fragment of metagenomic DNA coding for the enzymes that produce the drug(s) of interest, thus eliminating the need for complex sample tracking.

Our technology involves the transfer of drugs produced by metabolic pathways found in the metagenome into mammalian cells and into the tissues of model organisms such as nematodes and mice in a more direct manner.

In order to bypass some of the constraints normally felt with the drug screening of metagenomic libraries we came up with a method for better delivery which involves conferring an invasive property plasmid bearing bacteria carrying fragments of metagenomic DNA. In doing so, we are able to deliver bacteria producing natural products (drugs) directly into mammalian cells individually or 'en masse' for rapid, cost effective assays that bypass the limitations of traditional methods. Additionally, these invasive bacteria carrying libraries allow screening for drugs that affect tumor growth, longevity, obesity, etc. etc. in model organisms such as worms and mice. Drug screening with the use of bacterially delivered metagenomic libraries allows the isolation of novel drugs for a variety of in vitro as well as medicinal uses. The protocol for this invention:

1. Isolate metagenomic DNA or acquire inetgenomic DNA library containing fragments ~40kbp or greater (though fragments less than 40kbp could be used as well).

2. Transform library into bacterial strain of interest (e.g. attenuated E. coli) which has been modified to carry a bacterial invasion gene finv) which could come from Salmonella (inv 28) or another source microbe. This could also involve modification of the endogenous genome of the bacterial strain being used as a 'carrier' - as long as invasive properly towards target eukaryotic cells is conferred.

3. For in vitro mammalian cell assays - add to cell culture media or cells temporarily suspended in some media (e.g. saline solution). Allow invasion to occur over a period of time (e.g. 2hrs, 4hrs, overnight) and then add antibiotic to kill off remaining exogenous bacteria in the culture media. Then assay for phenotype (e.g. cell death) or readout (e.g. fluorescent reporter protein) of interest on a one- colony-per-well basis or 'en masse' and sort out cells considered to be positive hits. Specifically in the case of the en masse approach, cells deemed positive hits will be sorted (e.g. using FACS) as they will contain some bacteria carrying the metagenomic library clone of interest. A mild detergent (0.1% Triton X-100) will be used to "rescue" the internalized bacteria from the mammalian cells. Then, the initial assay will be repeated with the newly recovered bacteria, which should be enriched for plasmids carrying metagenomic DNA producing the novel drug or activity of interest. The process would be repeated as many times as deemed necessary until one bacterium or panels of bacteria producing the desired activity have been isolated.

4. For assays carried out in nematode, such as C. elegans, bacteria carrying metagenomic clones and having invasive properties will be plated in place of the standard bacteria used as a food source for C. elegans. To increase the sensitivity of the assay a one-metagenomic-clone-per-well format will be used. C. elegans in each well will therefore be given a continual dose of novel drug while a bacterially carried metagenomic metabolic pathway is being consumed. The assay may work for bacteria producing some novel drugs without having to confer the invasive property. In some cases the novel drugs that are produced may be hydrophobic and invasive bacteria may be needed to achieve cell membrane passage.

5. For assays carried out in mice, invasive bacteria carrying metagenomic clones will be will be orally administered in saline solution, similar to what has been done with invasive bacteria carrying plasmids expressing siRNA. To allow penetration of the cells of the intestinal epithelium a bacterial stain containing invasive properties will be needed. In other cases where an easily absorbed drug is produced, conferring an invasive property to the strain of interest (or directly utilizing an invasive strain such as Listeria) will be unnecessary.

The source of metagenomic DNA varies greatly and is approached in an unbiased manner or directed at a particular source that displays particular benefits. In an unbiased approach metagenomic DNA is isolated from local or distant soil or water samples or even a library of mixed sources. In the case of a focused library many options exist. In one embodiment, the library comes from a particular extreme environment such as the desert (e.g. bacteria that are resistant to DNA damage in areas with a very high UV index) or specific niches, such as water sources of a particular population of people/animals known to have low cancer incidence. Other sources involve isolating metagenomic DNA from sources such as the gut from populations of people or animals with low cancer incidence (e.g. amphibians such as frogs) or any pmlicular phenotype of interest. Screens from gut metagenomic DNA have the added potential benefit of isolating compounds that have a higher probability of being safe.

Lastly, in the case of in vitro screening of eukaryotic cells and in vivo screening and validation of hits in model organisms there is a tremendous potential for this technology to make disease treatment both affordable and more easily distributed than most current methods. Bacteria propagating a metagenomic DNA fragment can be frozen, or pelleted, desiccated and shipped anywhere. Multiple attenuated strains of bacteria would potentially be safe for consumption so that virtually unlimited amounts of drug could inexpensively be produced in the laboratory.

We here use our technology to isolate drugs or enzymes for any assay available or envisioned in mammalian or other eukaryotic ceils (e.g. protozoa, insect cells) that can be successfully infected with invasin or through any other protein, chemical or nanoparticle used to deliver bacteria intracellularly. It is a tremendous time saver (speed at which metagenomic libraries can be screened), space saver (no multiwall plates are necessary) and increases the sensitivity of screening (standard hit or miss methods of bacterial drug extraction unnecessary). Further, since the lysis of internalized bacteria happens over a period of several days we here carry out longer term assays without daily addition of a chemical library. Additionally, bacteria producing novel drugs of interest are here tested in higher organisms (e.g. C. elegans, mice) by feeding/oral delivery to examine effects of drugs on in vivo models of disease, without the need for costly large scale purification of compounds.

REFERENCES

1. Call, J.A., Eckhardt, S.G., and Camidge, D.R. 2008, Targeted manipulation of apoptosis in cancer treatment. Lancet Oncol 9(10): 1002- 101 1.

2. Chovan, T. and Guttman, A. 2002. Microfabricated devices in biotechnology and biochemical processing. Trends Biotechnol 20(3): 116-122.

3. Clausell-Tormos, J., Lieber, D., Baret, J.C., El-Harrak, A., Miller, O.J., Frenz, L., Blouwolff, J., Humphry, K.J.. Rosier. S., Duan, H. et al. 2008. Droplet-based microfluidic platforms for the encapsulation and screening of Mammalian cells and multicellular organisms. Chem Biol 15(5): 427-437.

4. Courtois, R, Olguin, L.F., Whyte, G., Theberge, A.B., Huck, W.T., Hollfelder, F., and Abell, C. 2009. Controlling the retention of small molecules in emulsion microdroplets for use in cell-based assays. Anal Chem 81(8): 3008-3016.

5. Davis, K.E.. Joseph, S.J., and Janssen, P.H. 2005. Effects of growth medium, inoculum size, and incubation time on culturability and isolation of soil bacteria. Appl Environ Microbiol 71(2): 826-834.

6. Ding, 7... Yang, X., Pater, A., and Tang, S.C. 2000. Resistance to apoptosis is correlated with the reduced caspase-3 activation and enhanced expression of antiapoptotic proteins in human cervical multidrug-resistant cells. Biochem Biophys Res Commun 270(2): 415-420.

7. Doi, N. and Yanagawa, H. 1999. STABLE: protein-DNA fusion system for screening of combinatorial protein libraries in vitro. FEBS Lett 457(2): 227-230.

8. Dumontet, C, Jordan, M.A., and Lee, F.F. 2009. Ixabepilone: targeting betalll- tubulin expression in taxane-resistant malignancies. Mol Cancer Ther 8(1): 17-25.

9. Elmore, S. 2007. Apoptosis: a review of programmed cell death. Toxicol Pathol 35(4): 495-516.

10. Gazitt, Y., Rothenberg, M.L., Hilsenbeck, S.G., Fey, V., Thomas, C, and Montegomrey, W. 1998. Bcl-2 overexpression is associated with resistance to paclitaxel, but not gemcitabine, in multiple myeloma cells. Int J Oncol 13(4): 839- 848.

11. Geng, H., Bruhn, J.B., Nielsen, K.F., Gram, L., and Belas, R. 2008. Genetic dissection of tropodithietic acid biosynthesis by marine roseobacters. Appl Environ Microbiol 74(5): 1535-1545.

12. Gillespie, D.E., Brady, S.F., Bettermann, A.D., Cianciotto, N.P., Liles, M.R., Rondon, M.R., Clardy, J., Goodman, R.M., and Handelsman, J. 2002. Isolation of antibiotics turbomycin a and B from a metagenomic library of soil microbial DNA. Appl Environ Microbiol 68(9): 4301-4306.

13. Goda, K., Bacso, Z., and Szabo, G. 2009. Multidrug resistance through the spectacle of P-glycoprotein. Curr Cancer Drug Targets 9(3): 281-297. rillot-Courvalin, (.'., Goussard, S., Huetz, Ojcius, D.M., and Courvaiin, P, 1998. Functional gene transfer from intracellular bacteria to mammalian cells. Nat Biotechnol 16(9): 862-866.

ess, DJ., Henry-Stanley, M.J .. Erickson, E.A., and Wells, C.L. 2003. Intracellular survival of Staphylococcus aureus within cultured enterocytes. J Surg Res 114(1): 42- 49.

Huebner, A., Srisa-Art, M.. Holt, D., Abell, C, Hollfelder, F„ deMello, A.J., and Edel, J.B. 2007. Quantitative detection of protein expression in single cells using droplet microfluidics. Chem Commun (Camb)(12): 1218-1220.

Huebner. A., Sharma, S., Srisa-Art, M., Hollfelder, F., Edel, J.B., and Demello, A.J. 2008. Microdroplets: a sea of applications? Lab Chip 8(8): 1244-1254.

Isberg, R.R., Voorhis, D.L., and Falkow, S. 1987. Identification of invasin: a protein that allows enteric bacteria to penetrate cultured mammalian cells. Cell 50(5): 769- 778.

Kim, J., Choi, J.N., Kim, P., Sok, D.E., Nam, S.W., and Lee, C.H. 2009. LC-MS/MS profiling-based secondary metabolite screening of Myxococcus xanthus. J Microbiol Biotechnol 19(1): 51 -54.

Kino, T., Hatanaka, H., Hashimoto, M., Nishiyama, M., Goto, T., Okuhara, M., Kohsaka, M., Aoki, H., and Imanaka, H. 1987. FK-506, a novel immunosuppressant isolated from a Streptomyces. I. Fermentation, isolation, and physico-chemical and biological characteristics. J Antibiot (Tokyo) 40(9): 1249-1255.

Knietsch, A., Waschkowitz, T., Bowien, S., Henne, A., and Daniel, R. 2003. Construction and screening of metagenomic libraries derived from enrichment cultures: generation of a gene bank for genes conferring alcohol oxidoreductase activity on Escherichia coli. Appl Environ Microbiol 69(3): 1408-1416.

Koster, S., Angile, F.E., Duan, H., Agresti, J.J., Wintner, A., Schmitz, C, Rowat, A.C., Merten, C.A., Pisignano, D„ Griffiths, A.D. et al. 2008. Drop-based microfluidic devices for encapsulation of single cells. Lab Chip 8(7): 1110-1115. Kurashina. R., Ohyashiki, J.H., Kobayashi, C, Hamamura, R., Zhang, Y., Hirano, T., and Ohyashiki, K. 2009. Anti-proliferative activity of heat shock protein (Hsp) 90 inhibitors via beta-catenin TCF7L2 pathway in adult T cell leukemia cells. Cancer Lett.

Lewis, K. 2007. Persister cells, dormancy and infectious disease. Nat Rev Microbiol 5(1): 48-56.

L Li, G., Wang, K., and Liu, Y.H. 2008. Molecular cloning and characterization of a novel pyrethroid-hydrolyzing esterase originating from the Metagenome. Microb Cell Fact 7: 38. Luo, C, Yang, X., Fu, Q„ Sun, M., Ouyang, Q., Chen, Y.. and Ji, H. 2006. Picoluer- volurae aqueous droplets in oil: electrochemical detection and yeast cell electroporation. Electrophoresis 27(10): 1977-1983.

Margulies, M., Egholm, M.. Altman, W.E., Attiya, S., Bader, J.S., Bemben, L.A., Berka, J., Bravenn n, M.S., Chen, Y.J., Chen, Z. et al. 2005. Genome sequencing in microfabricated high-density picolitre reactors. Nature 437(7057): 376-380.

Matz, C, Webb, J.S., Schupp, P.J., Phang, S.Y., Penesyan, A., Egan, S., Steinberg, P., and jelleberg, S. 2008. Marine biofilm bacteria evade eukaryotic predation by targeted chemical defense. PLoS ONE 3(7): e2744.

Paghdal, K.V. and Schwartz, R.A. 2007. Sirolimus (rapamycin): from the soil of Easter Island to a bright future. J Am Acad Dermatol 57(6): 1046-1050.

Prenen, H., Gil, T., and Awada, A. 2009. New therapeutic developments in renal cell cancer. Crit Rev Oncol Hematol 69(1): 56-63.

Rondon, M.R., August, P.R., Bettermann, A.D., Brady, S.F., Grossman, T.H., Liles, M.R., Loiacono, K.A., Lynch, B.A., MacNeil, I.A., Minor, C. et al. 2000. Cloning the soil metagenome: a strategy for accessing the genetic and functional diversity of uncultured microorganisms. Appl Environ Microbiol 66(6): 2541-2547.

Saleh-Lakha, S., Miller, M., Campbell, R.G., Schneider, K„ Elahimanesh, P., Hart, M.M., and Trevors, J.T. 2005. Microbial gene expression in soil: methods, applications and challenges. J Microbiol Methods 63(1): 1-19.

Sgro, A.E., Allen, P.B., and Chiu, D.T. 2007. Thermoelectric manipulation of aqueous droplets in microfluidic devices. Anal Chem 79(13): 4845-4851.

Shah, N.P., Nicoll, J.M., Nagar, B., Gorre, M.E., Paquette, R.L., Kuriyan, J., and Sawyers, C.L. 2002. Multiple BCR-ABL kinase domain mutations confer polyclonal resistance to the tyrosine kinase inhibitor imatinib (STI571) in chronic phase and blast crisis chronic myeloid leukemia. Cancer Cell 2(2): 117-125.

Shi, W., Qin, J., Ye, N., and Lin, B. 2008. Droplet-based microfluidic system for individual Caenorhabditis elegans assay. Lab Chip 8(9): 1432-1435.

Stevenson, B.S., Eichorst, S.A., Wertz, J.T., Schmidt, T.M., and Breznak, J.A. 2004. New strategies for cultivation and detection of previously uncultured microbes. Appl Environ Microbiol 70(8): 4748-4755.

Torsvik, V., Goksoyr, J., and Daae, F.L. 1990. High diversity in DNA of soil bacteria. Appl Environ Microbiol 56(3): 782-787.

Treusch, A.H., Kletzin, A., Raddatz, G., Ochsenreiter, T., Quaiser, A., Meurer, G., Schuster, S.C., and Schleper, C. 2004. Characterization of large-insert DNA libraries from soil for environmental genomic studies of Archaea. Environ Microbiol 6(9): 970-980. Tweddle, D.A., Pearson, A.D., Habcr, M., Norris, M.D., Xue, C, Flemming, (.'., and

Lunec, J. 2003. The p53 pathway and its inactivation in neuroblastoma. Cancer Lett 197(1-2): 93-98.

Vezina, C, Kudelski, A., and Sehgal, S.N. 1975. Rapamycin (AY-22,989), a new antifungal antibiotic. I. Taxonomy of the producing strcptomycetc and isolation of the active principle. J Antibiot (Tokyo) 28(10): 721-726.

Vogler, M., Butterworth, M., Majid, A., Walewska, R.J., Sun, X.M., Dyer, M.J., and Cohen, G.M. 2009. Concurrent up-regulation of BCL-XL and BCL2A1 induces approximately 1000-fold resistance to ABT-737 in chronic lymphocytic leukemia. Blood 113(18): 4403-4413.

Weigl, B.H., Bardell, R.L., and Cabrera, C.R. 2003. Lab-on-a-chip for drug development. Adv Drug Deliv Rev 55(3): 349-377.

Weinmann. L., Wischhusen, J., Demma, .J., Naumann, U., Roth, P., Dasmahapatra, B., and Weller, M. 2008. A novel p53 rescue compound induces p53-dependent growth arrest and sensitises glioma cells to Apo2L TRAIL-induced apoptosis. Cell Death Differ 15(4): 718-729.

Xiang, S., Fruehauf, J., and Li, C.J. 2006. Short hairpin RNA-expressing bacteria elicit RNA interference in mammals. Nat Biotechnol 24(6): 697-702.

Xue, C, Habcr. M., Flemming, C, Marshall, G.M., Lock, R.B., MacKenzie, K.L., Gurova, K.V., Norris, M.D., and Gudkov, A.V. 2007. p53 determines multidrug sensitivity of childhood neuroblastoma. Cancer Res 67(21): 10351-10360.

Zengler, K., Toledo, G., Rappe, M., Elkins, J., Mathur, E.J., Short, J.M., and Keller, M. 2002. Cultivating the uncultured. Proc Natl Acad Sci U S A 99(24): 15681-15686.