Nonviral methods are used to transfer different types of genetic material into cells. They are quite inefficient by comparison with viral methods, but because they are often quite simple and convenient methods, they have been widely used when the efficiency of transfer is not the highest priority. In the case of gene therapy, concerns about the safety of using virus vectors to transfer transgenes into the cells of a patient have also prompted interest in the alternative of nonviral transfer methods.
Physical methods
Physical methods are used to transfer genetic material into human or animal cells, either by piercing the cell membrane in some way, or by inducing pores in the mem brane that allow passage of the nucleic acids or oligonucleotides. Some methods have specialized uses. For example, microinjection of DNA, using a very fine needle to pierce the cell membrane, is limited to transfecting single cells at a time. A common application is to transfer DNA into fertilized oocytes (an important way of delivering genes into the germ line to make transgenic animals). Some of the more generally applicable physical transfer methods are described briefly below.
• Electroporation. When exposed to a sufficiently strong electric field, the plasma membrane of a cell undergoes electrical breakdown. Pores form, allowing pas sage of molecules that normally cannot cross the membrane (Figure 1). If the exposure is sufficiently short, the membrane can rapidly recover and become semi-permeable again. Electroporation (short for “ electric pore formation”) involves administering extremely brief pulses of very high voltage to the mem branes, allowing entry of desired large molecules and then resealing of the membrane.
• Sonoporation. An alternative way of transiently making pores in membranes to increase permeability is to use very brief pulses of high-energy ultrasonic sound.
• Particle bombardment. Biolistic methods use a gene gun to fire high-density (gold or tungsten) microparticles coated with nucleic acid; the microparticles are accelerated to very high velocity, usually using compressed gas, allowing efficient transfection of cells, irrespective of cell type. Developed initially to transform plant cells (whose cell walls pose a formidable barrier to passage of macromolecules), particle bombardment has been used to transfect plasmid recombinant DNA into a variety of cultured mammalian and animal cells, as well as tissues in vivo. The method is especially useful for transfecting cells that are more resistant to transfection by other methods.
Fig1. Electroporation as a way of permeabilizing the plasma membrane. Although the plasma membrane is a highly fluid structure, it is normally a formidable barrier to nucleic acids and oligonucleotides (which are negatively charged macromolecules). Electroporation involves exposing the cell membrane to very brief pulses of a high voltage electric field, causing pores to form transiently. Hydrophilic pores are bounded by the phosphate groups of membrane phospholipids and facilitate entry of nucleic acids or oligonucleotides into the cell.
Chemical methods
Certain chemicals can facilitate uptake of genetic material into mammalian cells by endocytosis. Calcium phosphate has long been used as an aid to gene transfer into mammalian cells. It relies on formation of DNA–calcium phosphate co-precipitates that are adsorbed onto the surface of target cells at high concentration, facilitating uptake by the cells through endocytosis. The transfection efficiency is not very high, however.
Various other chemical methods rely on cationic (positively charged) macromolecules that bind the negatively-charged nucleic acid or oligonucleotide molecules and target them to the cell membrane to facilitate their uptake into cells by endo cytosis. The vector–nucleic acid complexes have positively-charged surfaces and are attracted to cell membranes whose outer surfaces have numerous negatively charged phosphate and sulfate groups (within membrane-bound glycoproteins and membrane phospholipids).
Of the wide range of cationic vectors (Table 1), cationic lipids, as part of artificial lipid bilayers, have been especially widely used because of their comparatively high efficiency. This type of transfer (lipofection) uses synthetic spherical vesicles, known as liposomes, that have at least one lipid bilayer and form spontaneously when certain lipids are mixed in aqueous solution. After the desired nucleic acids or oligonucleotides are combined with a mixture of a cationic lipid and a helper lipid in water, cationic liposomes spontaneously form with bound nucleic acid/ oligonucleotide. After association with the cell membrane, they can be taken up into the cell by endocytosis (Figure 2). The efficiency may be increased still further when some other chemical vectors, such as polylysine, are also included in the mix.
Table1. EXAMPLES OF POSITIVELY-CHARGED CHEMICAL VECTORS FOR TRANSFERRING GENETIC MATERIAL INTO ANIMAL CELLS
Fig2. Receptor-mediated endocytosis and endosome maturation. (A) Receptor-mediated endocytosis. Different classes of protein on the cell surface act as receptors for specific ligands including signaling proteins and viruses. Invagination of the plasma membrane after ligands have bound to receptor proteins leads to formation of pits initially, and then small vesicles that transport the entrapped proteins or viruses within the cell. The pits are coated with a protein, such as clathrin, that plays a major role in vesicle formation, and forms a polyhedral lattice surrounding vesicles. Blue arrows indicate the direction of constriction as the plasma membrane invaginates. (B) General scheme for endosome maturation. Endocytic vesicles fuse with an early endosome, a sorting station composed of membrane limited tubules and vesicles. Some membrane proteins, such as receptor proteins, can be returned back to the plasma membrane for reuse, or be stored temporarily in recycling endosomes before being returned. Other membrane proteins and phagocytosed cells are transported via a multivesicular body to a late endosome; further sorting can occur that can lead to fusion with lysosomes and destruction of their contents. (A, adapted from Campbell NA & Reese JB [2008] Biology 8th edn. Pearson/Benjamin Cummings; B, from Alberts B et al. [2014] Molecular Biology of the Cell, 6th edn. Garland Science. With permission from WW Norton.)
Intracytoplasmic passage and nuclear entry
After passage through the cell membrane, transfected nucleic acids/oligonucleotides may need to escape from endosomes (if taken up by endocytosis). When taken up by endosomes, complexes containing bound nucleic acids or oligonucleotides might be expected to be quickly degraded: the endosome containing them would be shunted into the pathway that leads progressively toward formation of a late endosome and fusion with a lysosome (see Figure 2B). That fate can be avoided by designing the com plex to destabilize the endosome, allowing the transfected genetic material to escape. Figure 3 shows how that is achieved in the case of lipofection.
Fig3. Cationic liposomes as vectors for delivery of nucleic acids into mammalian cells. The nucleic acid to be transferred is complexed with liposomes to form lipoplexes that have positive charges on the surface, helping interaction with cell membranes (which have multiple negative charges on the surface). The lipoplexes are taken up by cells through different endocytosis pathways in which the cell membrane invaginates to form a pit. Large lipoplexes are taken up by pits coated with clathrin complexes (coated pit-mediated endocytosis at top right); small lipoplexes are taken up by noncoated pits (top left). In either case, the lipoplexes become trapped in endosomes (simplified here; see Figure 2B for a fuller picture) and would be expected to be targeted for destruction by lysosomes. However, the inclusion within the liposomes of certain helper lipids—usually electrically neutral lipids, such as dioleoyl phosphatidylethanolamine—helps to destabilize the endosomal membranes, causing the passenger nucleic acid to escape to the cytoplasm (yellow arrows). For a transferred DNA to be transcribed, it must pass to the nucleus. In dividing cells, the breakdown of the nuclear envelope during mitosis allows the DNA to gain access to the nucleus, but in nondividing cells the precise mechanism of entry into the nucleus is unclear. (From Simões S et al. [2005] Expert Opin Drug Deliv 2:237–254; PMID 16296751. With permission from Informa Healthcare.)
After escaping from the endosome, the genetic material is released from its protective chemical coat and becomes vulnerable to degradation by cytoplasmic nucleases (a defense system against invading viruses). To minimize degradation, oligonucleotides that are intended to work in the cytoplasm (to block RNA expression) are designed to be robust, nonstandard, synthetic oligonucleotides that are resistant to nucleases (Figure 4). For transgenes, the nucleus offers a safer environment, but whereas some virus vectors readily gain access to the nucleus, and even integrate into chromosomal DNA, using nonviral transfer to get a transgene into the nucleus is less straightforward (transport through the nuclear pores is generally inefficient).
Fig4. Chemical modification can increase oligonucleotide stability. The standard oligodeoxyribonucleotide structure is shown at the top. Four of the six modified oligonucleotides have a minor change, involving either replacement of atoms directly linked to carbons 2′ or 3′ of the deoxyribose (1,2) or replacement of an oxygen ion of the connecting phosphate (3,4). The other two modifications (5,6) are radical alterations to the normal structure, producing morpholino oligonucleotides or peptide nucleic acids (PNA). (Modified from Dias N & Stein CA [2002] Mol Cancer Ther 1:347–355; PMID 12489851.)
To facilitate nuclear targeting, various nuclear localization signal (NLS) peptides were developed to assist active transport of transgenes through nuclear pore complexes. Subsequently, however, nucleofection, a proprietary modification of the electroporation method using cell-type-specific reagents, was developed by the Amaxa company and has been used with considerable success to transfect transgenes into both the nucleus and cytoplasm. This method has been particularly useful for transfecting a wide variety of nondividing cell types, including neurons.
Transgene size range
As general methods for delivering genetic material to mammalian cells, nonviral transfer methods have two principal advantages over viral methods. First, they can readily transfer any type of genetic material—including short RNA molecules or chemically-modified oligonucleotides (viral methods are especially used to trans fer DNA, but viruses may convert DNA to RNA for propagation purposes). Secondly, nonviral methods can be used to transfer extremely large molecules: it has been possible to introduce transgenes containing megabases of DNA into human cells, where they replicate independently and behave as artificial chromosomes.
Viral methods of transferring DNA into mammalian cells
Over long periods of evolution, viruses have refined ways of packing their genomes into protective protein coats and injecting them into cells. According to the virus, the genome can be DNA or RNA and either single-stranded or double-stranded. Viruses are most readily manipulated as double-stranded DNA molecules to which a DNA of interest can be covalently attached, forming a transgene that can be packaged into a viral protein coat and transported into cells. In the case of RNA viruses, double-stranded DNA copies of the RNA genome (which naturally exist during the viral life cycle as replicative form DNA) are used. If required, the transferred DNA can be a copy of an RNA of interest (artificially made using a reverse transcriptase).
Viral transfer methods offer multiple advantages. First, they allow much higher trans fer efficiency than nonviral methods. Secondly, according to the type of virus, transgenes can be ferried into the cytoplasm or nucleus; in the latter case, retroviruses (RNA viruses that replicate through a DNA intermediate) allow integration of a transgene into the genome (retroviral integration into a host-cell chromosome is mandatory for successful completion of the life cycle). As a result, a transgene can shelter in the stable environment of chromosomal DNA and be inherited when cells divide. Additionally, certain strains of viruses are also suited to infecting particular types of cell.
Viral transfer methods do have some downsides. Although their protein coats protect the transferred DNA from nuclease attack, they impose size limits on the DNA that can be transferred—sometimes the maximum limit is just a few kilobases of DNA. And in the case of gene transfer in vivo there can be safety concerns. We will explore these in detail when we consider gene therapy in Chapter 22.
Transduction using retroviral vectors
A retrovirus has a single-stranded RNA genome but replicates in the host cell through the process of reverse transcription (in which the RNA is converted to DNA). A complete retrovirus particle (virion) has two copies of the RNA genome enclosed within a capsid protein coat; in turn, the capsid is surrounded by a lipid bilayer envelope with spike glycoproteins on the outside. In addition to different types of structural protein, the virion contains three key enzymes: a reverse transcriptase, a protease, and an integrase (Figure 5).
Fig5. Retrovirus structure. An infectious retrovirus particle (virion) has two copies of a single-stranded RNA genome, each with an m7GpppG cap at its 5′ end and a 3′ poly(A) tail (the genome is said to be a positive single-stranded RNA because in the cytoplasm the same RNA serves as a sense strand for making proteins). The genomic RNA is bound and structured by a nucleocapsid protein, and is enclosed within an inner capsid along with some viral enzymes (described in Figures 6 and 7). The capsid and its contents constitute the core particle, which is surrounded by an envelope consisting of a lipid bilayer with attached proteins. In addition to structural matrix proteins, the envelope periodically has spike glycoproteins consisting of an outer surface glycoprotein (which binds to specific receptors on the surface of cells) and a transmembrane glycoprotein (which aids virus entry into a cell by triggering fusion between the virus lipid bilayer and the cell’s plasma membrane).
Fig6. Retroviral life cycle. Infection begins when the virus envelope surface glycoprotein recognizes specific receptors located on the cell surface (1) and the virus enters the cell, usually as a core virus particle that lacks the outer envelope. Thereafter, the viral RNA is released from the capsid (2) and the viral DNA polymerase converts the single stranded (ss) RNA genome into a double-stranded (ds) DNA (3). That is possible because the viral DNA polymerase is multifunctional, having: a reverse transcriptase activity (uses the ssRNA to synthesize a complementary DNA—step 3a); an RNase H activity (degrades the RNA—step 3b); and a standard DNA polymerase activity (converts the ssDNA to a dsDNA—step 3c). After entry into the nucleus, the viral DNA inserts into chromosomal DNA using the viral integrase enzyme (4), enabling the viral genome (now called the provirus) to be stably maintained, replicated during DNA synthesis, and passed to progeny cells. Viral RNA is produced (5) and migrates to the cytoplasm (6) where it is translated to produce viral structural proteins and enzymes (7). Viral RNA also associates with some newly-produced viral proteins to form new core particles (8). The core particles then obtain their envelopes and are released from the cell (budding; step 9). Mature progeny virions are then capable of infecting new cells. LTR, long terminal repeat; R, short repeat in LTR (see Figure 7B).
Fig7. Functional components of a simple retroviral genome. (A) The positive single-stranded RNA genome in a virion has a 5′ cap and a 3′ poly(A) tail because the same RNA strand is also translated in the cytoplasm to make viral proteins. Each of the three genes makes a polyprotein precursor that is cleaved to give the individual proteins (see Figure 5 for the protein names and locations in the virus). The gag (group antigen) gene makes various structural proteins. The pol gene gets its name because one of its products is a multifunctional DNA polymerase: it can use both RNA templates—a reverse transcriptase (RT) activity—and also DNA templates. The pol gene also encodes a protease (PRO) and an integrase (INT), the enzyme used to insert the viral genome (as double-stranded DNA) into the host cell’s nuclear genome. The env gene makes the two envelope proteins present in the spike glycoproteins (see Figure 5). The retroviral protease cleaves the polyproteins encoded by gag and pol; cellular proteases are responsible for cleaving the env-encoded polyprotein. The flanking sequences contain a short repeat (R) plus some regulatory sequences, including U5 and U3 sequences (in the 5′ and 3′ untranslated regions, respectively) and psi (ψ), a crucial packaging signal that directs incorporation of the RNA genome into virions, which is located downstream of U5. (B) When the RNA genome is converted to double-stranded DNA, the R repeat is responsible for transfer of DNA synthesis between templates so as to cause duplication of the U5 and U3 sequences. As a result, the proviral genome has long terminal repeats, each containing a U5 and U3 sequence as well as the R sequence.
An infectious retrovirus particle enters a cell after its surface envelope glycoproteins bind to specific receptors on the cell surface (such as the CD4 receptor in the case of human immunodeficiency virus, HIV). Usually the transmembrane envelope glycoprotein then triggers the fusion of the viral and cellular membranes, allowing the virus to enter as a core particle, lacking the outer envelope; see Figures 5 and 6. (But some retroviruses enter by the receptor endocytosis mechanism described in Figure 2.) After infecting a cell, the core virus uses its multifunctional DNA polymerase to make a complementary DNA copy of its RNA, degrade the original RNA, and then copy the surviving single-stranded DNA (see Figure 6).
The resulting double-stranded DNA can be incorporated into the host-cell genome using the viral integrase. The integrated virus, known as a provirus, may remain in the host-cell genome and be transmitted to daughter cells following cell division. If germ line cells are infected, the virus may be transmitted vertically, but horizontal trans mission is the norm: after a virus infects a cell, the transcriptional and translational machinery of the host cell is hijacked to produce new copies of the viral genome plus viral proteins, after which new virus particles are assembled and then exit from the cell to infect other cells (see Figure 6).
Retroviruses have small genomes, from 7 to 10 kb in length. Simple retroviruses, such as gammaretroviruses, have three genes: gag, pol, and env. They each make protein pre cursors (polyproteins) that are cleaved to produce two or three individual proteins. In addition, various regulatory sequences are found in the flanking sequences (Figure 7). More complex retroviruses, including lentiviruses such as HIV, have extra genes involved in replication in addition to the gag, pol, and env genes.
Different classes of retroviral vector are used to transfer transgenes into cultured mammalian cells, but vectors based on gammaretroviruses, such as murine leukemia viruses, have been widely used. They cannot pass through nuclear pores, and so cannot be used with nondividing cells, but are able to access the nucleus of dividing cells after the nuclear membrane dissolves in preparation for mitosis and then re-forms. Initially, gammaretro virus vectors were commonly used in gene therapy, but safety concerns associated with their use have prompted the alternative use of lentivirus vectors for this application.
Standard recombinant gammaretrovirus vectors are made by using genetic engineering to modify a double-stranded gammaretroviral DNA, replacing retroviral sequences to create replication-defective vectors. The early steps in the retroviral life cycle—retroviral entry into cells, reverse transcription of the viral genome into DNA, and integration of the viral genome into the host genome—do not require viral protein. As a result, all viral coding regions can be deleted (and replaced by a desired transgene) and the remaining viral sequences can be reduced to the minimum required for high-efficiency transfer. The viral proteins do need to be supplied in trans and a packaging cell line is required to build a virus coat for a vector containing the desired foreign DNA. Virus particles obtained from such cells can then be used to introduce the desired DNA into a target cell of choice (Figure 8).
Fig8. Retroviral vectors are produced after first transfecting the recombinant retroviral DNA into a packaging cell line. (A) A transgene that is a recombinant retrovirus, often called a vector construct (VC). The gag, pol, and env genes have been deleted and replaced by a foreign DNA of interest, but the viral regulatory sequences have been retained. They include promoter/enhancer sequences, transcription termination signals, and the att (integration) and ψ (packaging) signals. (B) Packaging cell lines are prepared by transfecting a suitable cell with viral genes that can supply suitable viral proteins in trans to build a virus particle containing complementary RNA for a foreign DNA. For example, plasmids with the coding sequences for the gag, pol, and env genes plus nonviral upstream promoter/enhancer sequences (P) can be transfected into suitable mammalian cells, where they are stably maintained and where they produce viral structural and enzymatic proteins. When a retroviral vector is introduced into the packaging cell, vector construct RNA can be packaged, resulting in the production of virus particles containing a vector construct RNA genome. This virus can be harvested and then used to infect target cells to introduce foreign gene X in the vector construct into the cells, and to make gene X product. Because these target cells do not express viral proteins, the vector will not be propagated further. (The viral genes in the packaging cells are not carried along with the vector because they lack the cis-acting sequences necessary for propagation.)
Transduction using nonretroviral vectors
Various types of DNA virus have also been used to transduce mammalian cells, notably adenoviruses that normally cause benign infections of the upper respiratory tract in humans. The linear, double-stranded DNA genome remains nonintegrated within the cell nucleus, and unlike retroviruses they are able to infect both dividing cells and nondividing cells. So, in addition to transducing cultured mammalian cells, they can be used to ferry transgenes into nondividing cells in vivo.
For use as vectors, adenoviruses have two major advantages over retroviruses. First, the large genome size means that adenoviral vectors can accept quite large insert DNAs— “gutless” adenoviral vectors (which lack any of the adenoviral genes) can accommodate transgenes up to 35 kb in length. (A packaging cell line is required to provide viral proteins in trans, just as in the case of retrovirus vectors.) Second, adenoviruses can be produced in very high titers (much higher than retroviruses) and so they offer high levels of trans gene expression. Until recently, adenovirus vectors were widely used in gene therapy, but safety concerns have led to the use of alternative vectors, as described in Chapter 22.
Host range and tropism
Viruses gain access to cells after a protein on the outer virus surface recognizes and binds to a specific receptor on the host-cell surface. Different viruses bind different cell surface receptors, and even different strains of the same virus sometimes use different receptors. The virus (and virus vectors) may be limited to infecting cells from one species (the receptor is not so well conserved between species) or may be able to infect cells from a range of species (the receptor may have been very highly conserved during evolution). In the case of Moloney murine leukemia retroviruses, for example, the surface envelope glycoprotein encoded by the env gene is the protein responsible for receptor binding, but according to the strain of virus, the virus particles may infect murine cells only (the env gene is said to be ecotropic) or a range of mammalian species, including humans.
The tropism extends to cell type. For example, HIV is tropic for certain immune system cells that bear the CD4 receptor, notably helper T cells, macrophages, and dendritic cells. Certain strains of the same virus may preferentially infect cells of different types, as in the case of strains of adeno-associated virus.
