Patent Application
20240043818
On average there are 33 million musculoskeletal injuries in the United States per year, and fifty percent of these injuries involve tendons and ligaments (e.g., rotator cuff, anterior cruciate ligament) and are due to trauma or degeneration (Wu, et al., 2017). Tendons and ligaments are collagen-based soft tissues and are often subject to insufficient healing after injury. Further, the average reported recovery time for these tendon and ligament injuries is 12 weeks, and injured tendons may never fully return to the preinjury state (nhs.uk/conditions/hand-tendon-repair/recovery/). To speed up this lengthy recovery time, treatments are being developed involving implantation of type I collagen patches following surgical repair of these injured soft tissues. The surgical patches can help to bridge the gap between ends of torn tendons and can avoid stretching the injured tissue to keep it strong and to prevent reinjury (orthocarolina.com/media/how-does-a-patch-repair-a-rotator-cuff-tear). Even after these surgical interventions, the patient often does not regain the strength or mobility of the preinjury state. Currently the type I collagen for these patches is derived from animals and, consequently, such treatments present a significant risk of immune system rejection when implanted into human patients (Vig, et al., 2019). One example is the Zimmer© collagen repair patch, which is made for reinforcement of rotator cuff repairs. The patch is composed of an acellular collagen derived from porcine dermis (Yao, et al., 2005). The patch has shown promise over other biomaterial scaffolds, but its mechanical properties are not strong enough for tendon or ligament repair, and the manufacturing process used to sterilize and purify the patch compromises the collagen structure (OKeefe, et al., 2013). There is a need for improved sources of collagen for tissue repair and reconstruction in human patients.
The present technology provides engineered cells, including human cells, capable of greatly enhanced collagen production and methods of using them to obtain collagen for treatment of medical conditions without the risk of undesired immune reactions. Using the present methodology, cells can be obtained from a patient in need of collagen supplementation for treatment of a medical condition and then engineered to produce large amounts of the patient's own collagen for implantation, including auto-telocollagen, auto-procollagen, or auto-atelocollagen. Additionally, the method can be used with same species (e.g., human donor cells) to produce allo-telocollagen, allo-procollagen, or allo-atelocollagen.
The cellular engineering process overcomes bottlenecks in the collagen synthesis pathway via genetic engineering using CRISPR, together with optional use of chemical additives in the cell growth media that stimulate translation and post-transcriptional modifications involved in collagen synthesis. Genetic modifications can be made to human cells such as corneal fibroblasts using CRISPR activation (CRISPRa) to increase the expression of one or more genes responsible for the transcription, translation, and post-translational processing of collagen, including type I collagen. The collagen can be used, for example, in collagen patches for soft tissue repairs, and can be produced for a fraction of the current market cost for human collagen. Further, collagen provided by the present methods lacks telopeptide damage because there is no need for pepsin extraction. Because the collagen can be from a human source or from cells derived from the patient to be treated with the collagen, fewer screening and purification procedures are needed.
One aspect of the technology is an engineered cell capable of enhanced collagen biosynthesis. The cell has been engineered to perform CRISPR-based activation (CRISPRa) of a targeted gene related to collagen biosynthesis by the cell. As a result of transfection or transduction with components of the CRISPRa system, the cell expresses an endonuclease deficient Cas9 (dCas9) protein fused to a transcriptional activator protein (dCas9-activator) and also expresses a guide RNA (gRNA) specific for the targeted gene. The engineered cell is capable of increased collagen biosynthesis compared to a non-engineered cell of the same type. The collagen biosynthesis rate of the cell has been increased, such as, for example, by at least 10-fold, at least 20-fold, at least 30-fold, at least 40-fold, at least 50-fold, at least 60-fold, at least 70-fold, at least 80-fold, at least 90-fold, at least 100-fold, at least 200-fold, or at least 300-fold as a result of the CRISPRa engineering. The cell is preferably of a type that, as naturally occurring, produces the desired collagen type whose biosynthesis is activated by CRISPRa in the engineered cell. For example, the cell can be of a type selected from the group consisting of fibroblasts of any desired tissue or organ, chondrocytes, osteoblasts, epithelial cells, endothelial cells, mesenchymal cells, pericytes, hematopoietic cells, and fibrocytes.
In another aspect of the technology, the engineered cell described above has been engineered using CRISPRa to increase the expression of one or more genes selected from the group consisting of COL1A1, COL1A2, TGF-β1, TGF-β2, TGF-β3, COL2A1, COL3A1, COL4A1, COL4A2, COL4A3, COL4A4, COL4A5, COL4A6, COL5A1, COL5A2, COL5A3, ADAMTS2, ADAMTS3, ADAMTS4, ADAMTS5, ADAMTS6, ADAMTS7, ADAMTS8, ADAMTS9, ADAMTS10, ADAMTS12, ADAMTS13, ADAMTS14, ADAMTS15, ADAMTS16, ADAMTS17, ADAMTS18, ADAMTS19, ADAMTS20, TLL1, TLL2, and BMP1. Preferably, the cell has been engineered to increase the expression of 2 or more, 3 or more, 4 or more, or 5 or more of those genes in the same cell. For example, the cell can express gRNA molecules including the nucleotide sequence of any one or more of SEQ ID NOS:1-156.
In yet another aspect of the technology, the engineered cell has been engineered to perform CRISPRa of one or more further targeted genes selected from the group consisting of prolyl-3-hydroxylase family genes, prolyl-4-hydroxylase family genes, lysyl hydroxylase family genes, GLT25D1, GLT25D2, Grp78, Grp94, protein disulfide isomerase (PDI) family genes, calreticulin, calnexin, CypB, PPlase family genes, cyclophilins, FK506 binding protein (FKBP) genes, cyclophilin B (CypB), HSP47, TANG01, SEC13, SEC31, and Sedlin. The increased expression of these genes contributes to the post-translational processing or secretion of collagen from the engineered cell into the medium.
In still another aspect of the technology, the engineered cell has been engineered by CRISPRa to increase the expression of one or more collagen genes and one or more TGF β genes. Preferably, the cell also has been engineered to increase the expression of one or more propeptidase genes. Each of the activated genes can have its expression increased by at least 2-fold, at least 5-fold, at least 10-fold, at least 20-fold, at least 30-fold, at least 40-fold, at least 50-fold, at least 60-fold, at least 70-fold, at least 80-fold, at least 90-fold, at least 100-fold, at least 200-fold, or at least 300-fold. Each separate activated gene can be activated to a different degree, or to a similar degree, compared to other activated genes in the cell.
Regarding the aspect of the technology described above, the cell preferably can be engineered to increase expression of one or more collagen genes selected from the group consisting of COL1A1, COL1A2, COL2A1, COL3A1, COL4A1, COL4A2, COL4A3, COL4A4, COL4A5, COL4A6, COL5A1, COL5A2, and COL5A3. It also preferably can be activated to increase expression of one or more TGF-βgenes selected from the group consisting of TGF-β1, TGF-β2, and TGF-β3. The cell also preferably can be engineered to increase expression of one or more propeptidase genes selected from the group consisting of ADAMTS2, ADAMTS3, ADAMTS4, ADAMTS5, ADAMTS6, ADAMTS7, ADAMTS8, ADAMTS9, ADAMTS10, ADAMTS12, ADAMTS13, ADAMTS14, ADAMTS15, ADAMTS16, ADAMTS17, ADAMTS18, ADAMTS19, ADAMTS20, TLL1, TLL2, and BMP1.
The engineered cell preferably contains or produces mRNA capable of expressing a deactivated or “dead” Cas9 protein, lacking endonuclease activity, which is fused to one or more transcription activator proteins. For example, the transcription activators can be selected from the group consisting of VP64, p65, Rta (e.g., encoded by Epstein-Barr virus BRLF1), VPR (a combination of VP64, p65, and Rta), MS2, HSF1, and SAM (a combination of MS2, p65, and HSF-1). The activator SunTag also can be used to activate a gene and provide a fluorescent signal for transcriptional activation (Tannenbaum et al. (2014)).
Yet another aspect of the present technology is a cell culture containing any of the above described engineered cells. The cell culture can contain only a single cell type which are all identical, or it can contain a mixture of two or more differently engineered cell types, either derived from the same cell or type of cell, or derived from different types of cells. The culture can be derived from one or more cells obtained from a subject in need of collagen supplementation, thus providing a pathway to producing sufficient amounts of collagen that is genetically, biochemically, and immunologically identical to collagen of the subject's tissues. Preferably the cell culture is capable of at least 5, at least 10, at least 20, or at least 30 passages without substantial loss of collagen biosynthesis rate, or has been immortalized.
Still another aspect of the technology is a method for engineering a cell to provide enhanced collagen biosynthesis. The method includes the steps of: (a) providing a collagen-producing cell, a first nucleic acid molecule encoding a dCas9-activator, and a second nucleic acid molecule specific for a target gene related to collagen biosynthesis; and (b) transfecting or transducing the cell with said first and second nucleic acid molecules. The cell becomes capable of expressing said dCas9-activator and said gRNA, and the target gene is activated, whereby collagen synthesis by the cell is increased.
Yet another aspect of the present technology is a method of producing collagen. The method includes the steps of: providing the cell culture described above; growing the cell culture in a bioreactor under conditions in which collagen is biosynthesized by the cells of the cell culture; and harvesting and purifying collagen from the bioreactor. Preferably, the growth of the cells in the bioreactor is in the presence of a modulator or stimulator of collagen biosynthesis, such as an agent selected from acetaldehyde, ascorbate, hyaluronic acid, p-aminopropionitrile, transforming growth factor beta (TGF-β), insulin-like growth factor 1 (IGF-1), glutamine, and combinations thereof. Preferably, the cell growth medium of the bioreactor is concentrated after collagen biosynthesis by eliminating water and solutes having a molecular weight less than 50 Daltons while retaining higher molecular weight compounds, whereby propeptide cleavage of the collagen is accelerated.
Even another aspect of the technology is a kit of parts for engineering a cell to enhance biosynthesis of collagen by the cell. The kit includes: (i) a first nucleic acid molecule encoding a dCas9-activator protein; (ii) one or more second nucleic acid molecules specific for one or more target genes related to collagen biosynthesis; and (iii) optionally one or more reagents for transfecting or transducing a cell with the first and second nucleic acid molecules.
Another aspect of the technology is a medical device that contains an engineered cell, or cell culture, as described above, or collagen produced by such a cell or cell culture. The device can be an ex vivo device capable of producing collagen by engineered cells derived from a clinical subject's own cells. Alternatively, the device can be implantable in the subject's body, and can either contain engineered cells derived from cells of the subject, or can contain collagen produced by such cells; the collagen can be attached to a surface of the medical device, or can be contained in a reservoir for delivery into a tissue of the subject. The medical device can also serve as a collagen delivery device, whereby the device is disposed outside the subject's body, or is worn by the subject, and delivers collagen into a tissue of the subject's body.
Still another aspect of the technology is a method of treating a mammalian subject, such as a human, having a medical condition characterized by insufficiency of collagen. The method includes: obtaining collagen produced by a cell culture as described above, wherein the engineered cells of the culture are derived from the subject or from one or more other subjects of the same species; and administering the collagen to the subject. The collagen can be administered by injection into a tissue or placement within the body during surgery. The collagen can be in the form of a membrane, sheet, pad, solution, or gel, or contained within a cell scaffold, bandage, or wound dressing, or can be in the form of a coating on an implanted medical device. The medical condition can be, for example, a wound, a torn ligament or tendon, a bone fracture, damaged cartilage, an eye condition, a condition requiring cosmetic treatment or surgery, a dermatological condition, skin wrinkles or scars, or a burn.
The present technology can be further summarized by the following list of features.
As used herein, the prefix “auto” refers to a product (e.g., auto-procollagen, auto-telocollagen, or auto-atelocollagen) derived from cells of the same subject as the subject undergoing treatment using the product.
As used herein, the prefix “allo” refers to a product (e.g., allo-procollagen, allo-telocollagen, or allo-atelocollagen) derived from cells of the same species, but not the same subject, as the subject undergoing treatment using the product.
As used herein, “procollagen” refers to a newly synthesized, inactive collagen subject to activation by cleavage of propeptides from the procollagen.
As used herein, “telocollagen” refers to an active form of collagen, capable of assembly to form collagen fibrils, that is created by cleavage of propeptides from procollagen.
As used herein, “atelocollagen” refers to collagen stripped of its telopeptides, such as by pepsin digestion.
As used herein, the term “about” refers to a range of within plus or minus 10%, 5%, 1%, or 0.5% of the stated value.
As used herein, “consisting essentially of” allows the inclusion of materials or steps that do not materially affect the basic and novel characteristics of the claim. Any recitation herein of the term “comprising”, particularly in a description of components of a composition or in a description of elements of a device, can be exchanged with the alternative expression “consisting of” or “consisting essentially of”.
The present technology provides novel human cells engineered to increase collagen production. The cells are produced utilizing a CRISPR activation (CRISPRa) cellular engineering process to induce rapid collagen production from a variety of cell types, including human cells. Genetic modifications are introduced into cells that naturally produce a desired type of collagen using CRISPRa to increase the expression of one or more genes responsible for the transcription, translation, and/or post-translational processing of collagen. Collagen production from the engineered cells can be further stimulated by growing the engineered cells in the presence of one or more chemical additives in the cell culture medium to achieve even greater rates of collagen production. The synthesized collagen can be isolated, purified, and then used in collagen patches, gels, or other forms for soft tissue repairs.
The present inventors have achieved dramatically increased collagen production by applying cellular engineering to certain bottlenecks in collagen biosynthesis. Collagen has a unique protein structure. Collagen consists of three amino acid chains which form a triple helix. The primary amino acid sequence found in collagen is glycine-X-hydroxyproline or glycine-proline-X, where X is any other amino acid. The significant amount of glycine (every third amino acid) allows the helix to form in a tight configuration making the molecule structurally resistant to stress (Lodish, et al., 2000). Type I collagen molecules are 300 nm long and 1.5 nm in diameter. Each collagen molecule is composed of a characteristic right-handed triple helix which is composed of two alpha 1 chains and one alpha 2 chain. Each chain contains 1050 sequential amino acids. Once the individual collagen molecules are formed, they pack together side by side to form fibers with a diameter of about 10-300 nm and about a 67 nm gap between the “head” and “tail” of adjacent molecules (Schleip, 2012). N-terminal to C-terminal covalent bonds stabilize interactions of collagen molecules that are located adjacent to one another. The periodic pattern of molecule packing creates striations which are visible by electron microscopy. The links between molecules facilitate collagen packing stability to form strong fibrils. In addition to type I collagen, Table 1 lists examples of other collagen types and their features.
Collagen synthesis occurs primarily in fibroblasts through a process that spans both the intracellular and extracellular space. Type I collagen production is primarily controlled by two genes: Collagen type I alpha 1 (COL1A1), a strip of 17,533 base pairs on chromosome 17 that occurs after the 50,184,096th base pair, and Collagen type I alpha 2 (COL1A2), a strip of 36,671 base pairs on chromosome 7 that occurs after the 94,394,561st base pair (NIH, 2019).
Most genes responsible for collagen production contain an exon-intron pattern with an average number of exons ranging from 3 to 117. Depending on the type of cell and collagen, there are multiple transcription initiation sites and exon splicing mechanisms which result in different mRNA species (Gelse, et al., 2003). Specifically for type I collagen production, the pro-alpha 1 and pro-alpha 2 chain genes are transcribed from the COL1A1 and COL1A2 genes, respectively. During this phase of collagen production, the pre-mRNA undergoes both splicing and capping. The cellular transcription activity depends on cell type and is regulated by numerous growth factors and cytokines. Some of these growth factors include members of the Transforming Growth Factor Beta (TGF-β) family, fibroblast-growth factors, and insulin-like growth factors (Gelse, et al., 2003). The efficacy of these growth factors depends on the cell type.
Once translation has occurred, the collagen is in a pre-pro-polypeptide chain phase, and it moves to the lumen of the RER for post translational modifications (Wu & Crane, 2019; Lodish, et al., 2000). These molecules intrude into the lumen by the assistance of receptors that recognize the signal recognition domain of the collagen molecules (Gelse, et al., 2003). Three major modifications are made to convert this chain to procollagen. The first modification is the removal of the signal peptide on the N-terminal of the peptide chain by the enzyme signal peptidase. Efficient cleavage by the signal peptidase requires smaller amino acids (i.e., alanine, glycine, serine) just before the cleavage site, so that the signal peptidase 1 (SPase 1) can properly cleave the terminal (Tuteja, 2005).
The second modification is the hydroxylation, or addition of hydroxyl groups (—OH), of lysine and proline residues by hydroxylase enzymes (
The third modification is the glycosylation of hydroxylysine with glucose and galactose. During this modification, glucosyl and galactosyl residues are placed on the hydroxyl groups of hydroxylysine. Hydroxylation of specific proline and lysine residues (non-hydroxylated) in the middle of the chain are catalyzed by hydroxylysyl galactosyltransferase and galactosylhydroxylysylglucosyltransferase enzymes bound to the endoplasmic reticulum membrane. Oligosaccharides are also bound to asparagine residues in the C-terminal propeptide of procollagen (Lodish, et al., 2000).
After these three post-translational modifications are made, the glycosylated and hydroxylated chains assemble into a triple helix by folding, much like a zipper, as intrachain disulfide bonds are “zipped” together. The helix consists of two alpha 1 (I) chains and one alpha 2 (I) chain subunits. This assembly consists of three left-handed helices configured in a 1050 amino acid long right-handed coil, which forms from the C-terminus to N-terminus in the endoplasmic reticulum before further post-translational changes take place. C-propeptides also play a role in the assembly of the peptide chains into a collagen monomer (Gelse, et al., 2003).
After processing and procollagen assembly, the triple-helical molecule moves to the Golgi apparatus for final modifications and packaging inside the tubular portion of the complex known as vesicular tubular clusters (Wu & Crane, 2019; Bonfanti, et al., 1998). Within these clusters, the procollagen aggregates and is packaged within the Golgi compartment into secretory vesicles and released for transportation to the extracellular space.
Outside of the cell, collagen peptidase enzymes cleave the unraveled propeptides on the N-terminal and C-terminal to remove the ends of the molecule and convert the molecule to tropocollagen. The protease that performs the propeptide cleavage is procollagen C-proteinase. The tropocollagen terminates on both ends with telopeptides, which can be an issue in regard to antigenicity and immunogenicity (Stuart, et al., 1982; Lynn, et al., 2004). Collagen molecules have telopeptides on either side of their chains. The telopeptides do not form the typical triple helical formation and contain the amino acid hydroxylysine. Hydroxylysine residues form crosslinks at the C-terminal of one molecule and the N-terminal of two adjacent molecules (collplant.com/technology; Lodish, et al., 2000). These telopeptides also can be a source of immunogenicity if the collagen is transplanted into another species, or even intraspecies (Stuart, et al., 1982; Lynn, et al., 2004; Uchio, et al., 2000). The triple helical region of collagen is conserved across species. Although variations in amino acid sequences within the helix differ by less than a few percent between species, up to fifty percent of the amino acid sequence in the telopeptides can differ between species (Lynn, et al., 2004). Due to this high interspecies variation in this region of the molecule, telopeptides are thought to be the primary contributing factor to immune responses post collagen implantation.
The final extracellular step is fibrillogenesis. Fibril-forming collagen molecules spontaneously self-assemble into ordered fibrillar structures. Long thin collagen fibrils are formed when lysyl oxidase covalently bonds lysine and hydroxylysine molecules. This behavior is encoded in the collagen structure. Fibril orientation depends on the type of tissue (Gelse, et al., 2003). Each fibril has a diameter of about 100 nm after the molecules are packed together side by side, although fibril diameter can range from 25-500 nm.
Clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR associated protein-coding genes (Cas) are a group of proteins used by the immune system of prokaryotes. CRISPR associated protein 9 (Cas9) can cut DNA, and is a highly efficient DNA targeting enzyme that has been modified for gene editing research applications. The CRISPR-Cas9 system is made up of four main parts: the Cas9 enzyme, guide RNA (gRNA), protospacer adjacent motif (PAM) sequence, and matching host DNA (matching genomic sequence). Cas9 is an endonuclease enzyme that utilizes an approximately 20 base pair section of guiding RNA to recognize, unzip, and induce double-strand breaks in DNA (Biolabs, 2019). Guide RNA (gRNA) directs the CRISPR-Cas9 system where to go in the genome and can result in the process of CRISPR-Cas9 cutting the host DNA, and then letting natural DNA repair processes incorporate an inserted gene of interest into the host's genome at a very particular point in the host genome (as defined by the gRNA). The gRNA has two main components, CRISPR RNA (crRNA) and trans-activating crRNA (tracrRNA) (
The matching host DNA is a 18-23 base pair sequence that is in or near the promoter region of the gene of interest. This sequence must be immediately followed by an NGG (representing any base pair followed by two guanines) or ‘PAM sequence’. In order to determine what base pairs should be targeted by the gRNA in a given gene, computational biology tools have been developed to find PAM sequences and base pair sequences of interest (CRISPR Guide Design Software, Pelligrini, 2016). These base pairs are ranked for how well the Cas9 system is able to bind to that sequence without accidentally attaching to other similar sequences in the genome. Additionally, manufacturers, like Dharmacon™, have proprietary software that is used to compute highly specific binding sites for gRNA.
The present inventors elected to use CRISPR gene activation (CRISPRa) to engineer cells that produce much greater amounts of collagen than naturally occurring cells. In CRISPRa, a deactivated or dead Cas9 enzyme (dCas9) without endonuclease activity is used together with a guide RNA (gRNA or sgRNA) to locate to a specific gene target. The dCas9 can be fused to one or more transcriptional activator proteins. The resulting fusion protein is referred to herein as “dCas9-activator”. The one or more activators fused to dCas9 can be, for example, VP64 (a tetramer of the Herpes Simplex Viral Protein 16) or VPR (VP64 bound to β53 and an R transcription factor)). For example, the VPR activator can be dCas9 from S. pyrogenes fused to VP64-β65-Rta. Other activators or combinations of activators can be selected according to cell type or gene to be activated. The dCas9 does not cut the bound DNA, but acts to upregulate expression of the targeted gene. The activator domain fused to the dCas9 causes transcriptional activation by recruiting transcription complexes to the promoter regions of these genes.
An important design consideration when performing CRISPRa is selecting where dCas9 binds to the gene; generally, a position on the gene's promoter is selected. While specificity of a gRNA sequence and locations of its PAM sequences can be predicted using computer algorithms, the location on the promoter and its resulting effectiveness is variable. While the promoter area for effectiveness of CRISPRa is generally 50-400 base pairs upstream of the transcription start site, the most effective location for activation varies between genes, and some locations are completely ineffective (Mohr, et al., 2016).
There are two practical ways to deliver the CRISPR-Cas9 system for performing CRISPRa (i.e., the dCas9-activator and gRNA for genes to be activated) into a cell: transfection and transduction.
Transfection is the delivery of nucleic acids (typically the mRNA corresponding to the transcribed gene) into a cell and subsequent translation of the mRNA by the host cell. When performing CRISPRa using transfection, the CRISPR-Cas9 system is usually expressed for about 24-48 hours. Transfection of cells with the CRISPRa components can be performed by microinjection, electroporation, or use of ribonucleoprotein (RNP) complexes to deliver the mRNAs. Transient expression using transfection is simpler and less expensive than transduction, and decreases the odds of off-target activation due to its short expression window. Further, the use of mammalian expression vectors allows for transfection that is less transient than traditional, non-mammalian transfection vectors.
Commercial kits are available for performing CRISPRa by transfection of cells. For example, Dharmacon (horizondiscovery.com) offers a protocol and reagents for pooled transfection of gRNA and dCas9 mRNA for culture in a 96-well plate using the DharmaFECT Duo Transfection Reagent. The protocol can be scaled up to a 48-well plate or further in order to harvest more collagen, such as for more accurate collagen quantification while engineering cells for increased collagen production. Additionally, when two or more different sets of gRNA are used to activate two or more genes simultaneously, appropriate adjustments can be made to the amount of dCas9 mRNA and the amount of transfection reagent. For example, in order to simultaneously activate expression of the COLA1, COL!A2, and TGF-β3 genes in a human cell, the materials can include CRISPRa Human COL1A1 crRNA pool, CRISPRa Human COL1A2 crRNA pool, CRISPRa Human TGF-β3 crRNA pool, CRISPR-Cas9 Synthetic tracrRNA, Edit-R GFP dCas9-VPR mRNA, DharmaFECT Duo Transfection Reagent, 10 mM Tris-HCl Buffer pH 7.4, and serum-free medium. Example reagents needed for a pooled transfection of human corneal fibroblasts for increased collagen production are shown below in Table 2. Only one pooled crRNA purchase is necessary as target crRNAs can be mixed and matched in one pool purchase (one pool contains four times the minimum single crRNA needed for CRISPRa), but having more than the minimum crRNA necessary can lead to better gene activation (CRISPR Guide Design Software).
The CRISPRa Human COL1A1 crRNA pool includes a pool of individual RNA sequences complimentary to different regions of the COL1A1 promoter (see target sequences SEQ ID NOS:1-4 in Table 3).
The DharmaFECT Duol Transfection Reagent has been shown to be an efficient transfection reagent for transfection of small RNAs and plasmids simultaneously (Borawski, et al., 2007).
An example of a dCas9 protein for use with the present technology is one having the amino acid sequence shown below (SEQ ID NO: 157 (uniprot.org/uniprot/AOA386IRG9)):
An example of a synthetic tracrRNA for use with the present technology is one published by Jinek, et al., (2012), which has the sequence GGAACCAUUCAAAACAGCAUAGCAAGUUAAAAUAAGGCUAGUCCGUUAUCAACUUGA AAAAGUGGCACCGAGUCGGUGCUUUUUUU (SEQ ID NO:158). The crRNA may include a region complementary to a portion of the tracrRNA. Alternatively, a linker sequence can be added between the crRNA and the tracrRNA to yield a single gRNA molecule.
VP64 is a transcriptional activator including four tandem copies of VP16 (Herpes Simplex Viral Protein 16, amino acids 437-447, connected with glycine-serine linkers). The amino acid sequence of VP64 is shown below (SEQ ID NO: 159, (parts.igem.org/Part:BBa_J176013)):
The transcriptional activator β65 includes four isoforms produced by alternative splicing (uniprot.org/uniprot/Q04206). Isoform 1 has the amino acid sequence shown below (SEQ ID NO: 160).
The amino acid sequence of the transcriptional activator HSF1, SEQ ID NO: 161, is shown below (SEQ ID NO: 161 ((uniprot.org/uniprot/Q00613)).
The amino acid sequence of transcriptional activator MS2 is shown below (SEQ ID NO: 162 (uniprot.org/uniprot/P03612)).
The dCas9 mRNA, described above, is the limiting reagent for the protocol described below, allowing for 11 wells to be made with 5 nmol of starting material. An example protocol for plating the test plate conditions is:
To measure collagen production, a hydroxyproline assay can be used, optionally after concentrating the collagen solution or culture medium (see, e.g., D.D. Cissell et al., (2017). Tissue Eng Part C Methods. 2017 Apr;23(4):243-25) Alternatively, the SirCoI™ dye binding collagen assay can be used (www.biocolor.co.uk/product/sircol-soluble-collagen-assay). A standard curve, as shown in
Three levels of transfection reagent were compared; transfection reagents can be toxic to cells at high concentrations. The high (H), medium (M), and low (L), transfection reagent concentrations were 7.2, 4.05, and 0.9 μL of transfection reagent per well, respectively. Media was harvested every 12 hours and pooled in two-day segments. Collagen amount per well was quantified every two days, and cell counts were performed after the end of media harvesting. Collagen produced per cell was estimated based on the number of cells per well at the end of media collection. A comparison of total (days 1-6) pooled collagen production (per well, not per cell) for each concentration level and the cell count is shown in
Table 5 shows mean collagen production (molecules/cell/hours) for the test groups (high, medium, low), and the group names denote the transfection level and the days at which the assay was taken (i.e., 1-2=days 1 and 2).
Quantifications that fall below the minimum detectable collagen amount are set to the minimum detectable collagen amount (1.06 million molecules/cell/hour) (Table 5). Having some quantifications fall below the minimum detectable amount is necessary to capture the upper limits of collagen production in the CRISPR conditions. Data for days 1-2 and most control samples were at or below the minimum detectable collagen amount, but every datapoint for CRISPR results at days 3-4 and 5-6 is above the minimum threshold. The fact that the results at days 1-2 matched the control wells is consistent with the expectation that CRISPR should upregulate collagen production after about two days. This increase in collagen is shown in a graphical format in
Based on the graph in
Table 6 shows the subset groupings for the experiment represented in
In order to prevent undesired crosslinking of newly synthesized collagen, BAPN is preferably added to the culture medium. Culture conditions may also be adjusted to optimize preservation of collagen structure and function. For example, since collagen is unstable if stored at 37° C., cells can be cultured below 37° C., and/or collagen can be harvested periodically (e.g., every 8-24 hours), and then stored at a lower temperature to preserve stability.
From the data in Table 5 and
As discussed above, it was found that the CRISPR system can increase collagen production about 90.29-fold in collagen production compared to the experimental control. When scaled up to a T-75 flask, this production level would yield about 554 mg collagen/week. However, the purchasing of commercially available dCas9 mRNA, gRNA (crRNA and tracrRNA) reagents for each gene of interest, and transfection reagent, would be cost prohibitive. Therefore, it is desirable to scale up using other methods of CRISPRa delivery that would cost significantly less. Suitable methods include the use of bacteria to produce dCas9 and gRNA plasmids and viral vectors produced by Human Embryonic kidney (HEK) cells to deliver and stably integrate the dCas9 and gRNA sequences into the host genome would significantly reduce the cost of scale-up and result in lower collagen production costs.
A CRISPRa process involving transduction of cells using a lentiviral vector is illustrated in
For transduction, once the virus production pipeline is established, viral plasmids are cheaply generated by bacteria, transfected into HEK cells that readily package and excrete a supply of viral particles, and added to host cell media for integration of gene of interest into host cell's genome.
The first decision to make for scaling up the collagen production process with viral delivery is deciding what kind of virus to use. Briefly, the benefits of using AAVs (adeno-associated viruses) include high titers, versatility from the availability of multiple serotypes that target different cell types, low toxicity as the virus remains in the episome or in a specific locus on chromosome 19, and low immunogenicity as there is minimal host immune response. The benefits of LVs include that they infect nearly all mammalian cell types, they can be used to deliver relatively large DNA sequences-usually about 5-6 kb in length, and they can be used to generate stable cell lines or drive stable gene expression in organs and tissues in vivo due to integration of the transgene at random locations in the genome. Because the LVs allow for quick and easy stable integration of transgenes, they are a clear choice here, especially because the same class of LV could be used for any cell type that is chosen to work with (epidermal fibroblasts, corneal fibroblasts, iPSCs, etc.). AAVs might be considered if there was a desire to develop a system to be used in a patient that targets a specific tissue type for increased collagen production. The details of the components that make up a LV and how plasmids can be used to construct LVs are further detailed herein in the example shown in
There are three main genes involved in lentivirus construction: env, gag, and pol. Env, or recombinant VSV-G, (
The last component is the transfer plasmid, which carries the vector of interest to be integrated into a host cell. This vector is usually engineered with restriction enzymes before being inserted into a bacterial cell for replication. As such, the transfer plasmid has a viral region with the vector of interest (and some other needed components) and a non-viral region with an antibiotic resistance gene. This antibiotic resistance is used for selection against bacteria without the new engineered plasmid following transformation (plasmid insertion). The viral region is set off by the aforementioned LTRs. In addition to acting as a promoter once in the host genome, these LTRs on either side of the vector of interest allow the vector to be recognized by multiple other viral proteins in the virus production and transduction process.
Once in the host genome, transcription of this vector has two main functions. The first is to select against cells that have not been transduced. In the transgene there is an antibiotic resistance gene that is transcribed by host cells (
A lentiviral vector for expressing dCas9-VPR is commercially available (Dharmacon Edit-R CRISPa Lentiviral dCas9-VPR). This can be used together with another vector encoding the gRNA specific for the target gene, either as a single guide RNA molecule or as separate crRNA and tracrRNA molecules complexed together to form a guide RNA molecule.
Research has demonstrated that using glass as a substrate for fibroblast cells yields more collagen production than in culture. When cells are placed on a substrate that is not encountered in vivo, they produce collagen as a method of isolating themselves from the plate.
Fibroblasts also have a tendency to spread out collagen across a substrate of a higher stiffness, such as glass, rather than clumping. Following this, collagen layers would begin to stack under the cells on the plate. In most collagen producing studies, the amount of collagen produced is limited by the substrate size.
The apparatus includes a series of stacked glass slides constrained on either fluid inlet/outlet end by jigs and watertight side walls on non-flow sides (
Following flow shearing, the culture and collagen fibrils can be removed from the glass plates and the amount of type I collagen could be assayed to determine how much was produced.
An alternative method of collagen growth using glass as a substrate is to utilize glass beads. An example device is depicted in
Three levels of transfection reagent were compared for a CRISPR design. The transfection reagent was DharmaFECT 1 Transfection Reagent as shown in Table 2. In all cell culture media, the media should be concentrated at least 10 times using a dialysis (50 Daltons molecular weight cutoff) against PBS to promote propeptide cleavage. The high, medium, and low, transfection reagent concentrations were 7.2, 4.05, and 0.9 μL of transfection reagent per well, respectively. Media was harvested every 12 hours and pooled in two-day segments. Collagen amount per well was quantified every two days, and cell counts were performed after the end of media harvesting. Collagen produced per cell was estimated based on the number of cells per well at the end of media collection. A comparison of total (days 1-6) pooled collagen production (per well, not per cell) for each concentration level and the cell count is shown in
Quantifications that fell below the minimum detectable collagen amount were set to the minimum detectable collagen amount (1.06 million molecules/cell/hour). Having some quantifications fall below the minimum detectable amount was necessary to capture the upper limits of collagen production in the CRISPR conditions. In other words, we could have not diluted the samples before quantifying collagen, but then the upper limits of CRISPR collagen production would have been above the upper threshold for collagen quantification and not have been detectable. Note that days 1-2 data and most control samples were at or below the minimum detectable collagen amount, but every datapoint for CRISPR results for days 3-4 and 5-6 were above the minimum threshold. This is important for when fold change and statistical significance was calculated, as the actual collagen production for the control wells is likely much lower than what we are saying it is. Lastly, early timepoints (days 1-2) matching the control wells makes sense as the CRISPR was expected to upregulate collagen production after about two days. Based on the graph in
An important feature of
The peak of collagen production seen using CRISPR showed about a 90.29-fold increase (42.1% std) in collagen production compared to the experimental control.
Cells that have been genetically modified with CRISPR can be cultured in media containing the top performing chemical conditions found in Example 2, including a control with CRISPR cells in traditional media and a second control.
In addition to the scale up methods depicted in
A variety of chemical stimulants were studied to be potentially used to increase the collagen production of fibroblast cells. Acetaldehyde, also known as ethanal, is a derivative of ethanol that has been shown to increase the levels of collagen produced in baboon liver myofibroblasts and human dermal fibroblasts when added to the culture media in concentrations up to 300 μM.
Ascorbic acid is known to be beneficial to the production of collagen in cell types including bovine, mouse, and human. It was hypothesized that the addition of caffeine to media could be beneficial to any lentiviral based CRISPR design solutions because of its demonstrated effect on increasing the activity of lentivirus in the gene therapy space. However, caffeine was shown to have a negative effect on collagen production in concentrations in media as little as 1-5 mM. This is most likely due to the inhibition of the accumulation of several growth factors, including interleukin-8.
While in high concentrations the addition of ethanol to cell culture media is commonly known to be extremely negative, a number of studies have been performed at concentrations of 50 mM in order to hypothesize the cause of alcoholic liver fibrosis. While fibrosis is commonly synonymous with an increase in the amount of collagen present in tissue, studies performed at these concentrations in myofibroblasts and fibroblasts report no direct effect on the level of collagen production.
Glutamate has not been studied extensively for its impact on collagen production despite its close relationship to glutamine, which has been known to be a very positive influence on collagen transcription rates. One study showed a 400% increase from the control in glutamate-supplemented media used to culture human dermal fibroblasts.
Many studies have been conducted in order to observe the response of human dermal fibroblasts in the presence of glutamine in media. Concentrations seen in these studies range up to 10 mM, demonstrating a maximum benefit for the production of collagen of nearly 300% of the control at 250 μM. This effect has been thought to increase the level of collagen gene transcription through its conversion into pyrroline-5-carboxylate.
Hyaluronic acid showed a neutral effect on collagen production in human dermal fibroblasts plated on an undisclosed surface at 500 pg/ml in media. It increased the rate of cell division and general fibril production when present at 150 pg/ml in media for human dermal fibroblast cultures plated on collagen, however researchers did not specifically quantify the production of collagen. When it was incorporated into the culture surface at a ratio of 1:19 hyaluronic acid to collagen by weight, it increased the amount of collagen produced by embryonic chick fibroblasts up to 230% of control. Based on the information presented in these studies and the well-known tendency of cells to stop producing collagen once enveloped in it, the group hypothesized that it interferes with a feedback process, fooling cells into making more collagen than they actually require for a suitable microenvironment.
There has been a demonstrated positive correlation between the level of Insulin-like Growth Factor 1 (IGF-1) in media and the production of collagen by manipulation of rat serum applied to media used to culture human dermal fibroblasts. Other studies quantify these values in human lung fibroblasts at a maximum of a 300% increase from control at a concentration of 100 ng/mL. Macroscopically, this effect can be seen in diabeteic individuals who are slow to heal wounds or suffer from accelerated atherosclerosis.
Interleukins (IL) encompass a wide range of glycoproteins associated with immune response. Researchers have looked into types 1p, 4, 6, 8, 10, and 13 for their specific effect on collagen production. IL-4 has demonstrated a maximum positive effect of a 250% increase from control. The lowest concentration of any of these types that is needed for an observable, positive effect is IL-1p at 2.5 μM in human chondrocytes. Out of the six types listed above, only IL-10 was found to reduce the level of collagen production.
Lactate is commonly found in high concentrations in the body after alcohol consumption, especially in individuals suffering from alcoholic liver fibrosis. Therefore, its addition into media could lead to an increase in collagen production even outside of the whole organ system. One study seems to agree with this logical argument by demonstrating a statistically significant increase in collagen production upon addition of 5 mM to media used to culture baboon liver myofibroblasts. However, in human dermal fibroblasts a concentration of 40 mM was shown to decrease collagen production.
Lathyrogens have been used to inhibit the formation of collagen crosslinks without cytotoxic effects. The most popular lathrogen used in cell culture is β-aminopropionitrile (BAPN) which operates by irreversibly blocking lysyl oxidase. Other cellular effects include prevented development of adhesive strength and a buildup of GuHCI-extractable collagen crosslink precursors. No research with any cell type has shown adverse effects on cell viability, collagen synthesis, or non collagen protein synthesis. However, one research study demonstrated inhibited fibroblast migration in a dose-dependent fashion at 0.25 and 0.5 mM BAPN. Previous research has used BAPN successfully at concentrations of 0.1 mM-0.5 mM.
Proline stabilizes collagen during post translational modifications. One study found a range of concentrations from 5-10 mM in media applied to human dermal fibroblasts that resulted in a maximum increase of 200% in collagen production as compared to the control value.
The effects of pyrroline-5-carboxylate on collagen production have been documented multiple times in human dermal fibroblasts. Information presented in these studies suggests an optimal concentration of 1 mM for a maximum increase of 260% of the control value. Interestingly, this effect can be seen in as little as 6 hours. It is thought to have such a potent effect because it enables IGF-1. Additionally, it can be converted to proline.
The subfamilies within the TGF-β family all have been shown to have a positive impact on collagen production. Generally, it has a positive effect on the number of ribosomes in the cell, the organelle responsible for the translation of all proteins, including collagen. TGF-β1 applied to rat liver M cells at a concentration in media as low as 1 ng/ml demonstrated an unquantified increase in collagen production from control. It has been shown that types of collagen produced by TGF-β vary, with collagen type I being especially associated with TGF-β3. Sources seem to agree on a concentration of 12.5 ng/ml for maximum efficacy in human dermal fibroblasts.
Based on a weighted scoring (Table 13), the seven best additives were selected and were added to the fibroblast media separately in a Phase 1 screening study using concentrations presented in Table 14. Table 14 shows the additives and concentrations to be tested. A common concentration cited in literature should be a standard condition, plus one concentration at 50% of that value, and another concentration at 150% that value (Table 14). Positive effects of these additives have been shown in other cell lines, but its main effects on corneal fibroblasts need to be explored independently before combinations of additives can be tested. This screening of factors phase is common in designed experiments in bioengineering applications. Specifically when using a factorial design for media composition it is recommended to perform an screening experiment of the unknown domain before applying advanced designs that allow optimization. Screening studies can be done in a number of ways. The simplest screening experimental design is each variable at two levels, however this assumes a linear relationship between the input and output. For this study a three-level design was chosen in order to determine a maxima. This phase of the experiment determined which additives have a statistically significant effect on collagen production and what concentration of each additive has the highest positive impact. This data was fed into Phase 2 of the experiment where the top performing additives were used within their optimal range of concentrations.
The seven additives also allowed for minimizing the number of plates and maximizing the number of used wells. This resulted in 3 plates, each with a standard media control group. In phase 1 of testing 7 media additives were tested at 3 concentrations with a sample size of 4 for each condition. Additives and concentrations used were determined by the prior research described above. Collagen amount per well was quantified, but cell counts were not performed. Collagen produced per cell was estimated based on the number of cells expected per well. The results are shown in Table 15. Some groups fell below the SirCol standard curve range of 1-50 ug. These values were represented as 1.00 ug/100 uL and were assigned the lowest rating for efficacy in the resulting trade study.
Three different top performing additives were identified (BAP, ACE, and ASC). Based on single additive performance, with only three media additives resulting in higher collagen production than the control, the three different additives went on to the DOE. For the three additives that advanced the highest ranking of each additive was used as the DOE centerpoint. Due to a lack of available stock, ACE low was substituted for ACE medium. As a result of these modifications, the final selections from Phase 1 were BAP low, acetaldehyde low, and ascorbate low for the centerpoint and all three at medium concentration for the “high” point (100%).
In phase 2 the top performing media additives were fed into a full factorial DOE. The concentrations used were determined by a variety of factors including a trade study, remaining laboratory supplies, and finally the lower cost associated with a lower concentration in a near-tie situation. A near-tie situation was defined by the group as concentrations of the same chemical that scored within 1% total score. For BAP, the concentration in the centerpoint was determined to be the low concentration from phase 1 at 0.25 mM due to it scoring the highest in the trade study. For ascorbate, a near-tie situation was observed so the low concentration from phase 1 at 0.5 mM was chosen. For acetaldehyde, due to a stock issue the low concentration at 0.2 mM was chosen. For the 100% concentration limit needed to complete construction of the full-factorial DOE plan, each of these values was doubled, which corresponds to the medium conditions from phase 1.
As shown in
OKeefe, R. J., Chu, C. R., Jacobs, J. J., & Einhorn, T. A. (2013). Orthopaedic basic science: foundations of clinical practice (4th ed.). Rosemont, IL: American Academy of Orthopaedic Surgeons.
This application claims priority to U.S. Provisional Application No. 63/092,433, filed 15 Oct. 2020, which is incorporated by reference herein in its entirety.
This invention was made with government support under Grant No. EY029167 awarded by the National Institutes of Health. The government has certain rights in the invention.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/US2021/055315 | 10/15/2021 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 63092433 | Oct 2020 | US |
