Immunizing Animals

Choice of Animal

The choice of animal for immunizations is determined by four points: (1) whether polyclonal or monoclonal antibodies are desired; (2) how much serum is needed; (3) from which species the antigen is isolated; and (4) how much antigen is available.

Whether to generate a polyclonal antibody or a monoclonal antibody largely depends on its subsequent application. Polyclonal antibodies are produced from many different B cells, each secreting a unique antibody. Polyclonal antibodies react with multiple epitopes on an antigen. Monoclonal antibodies, by definition, are a single subclass of antibody originally derived from a single B cell that recognizes a single epitope on an antigen. This single B cell is immortalized into a hybridoma cell line by fusing it with a myeloma cell. Polyclonal and monoclonal antibodies have many overlapping uses, but some applications are more suited to one or the other (Tables 1, 2).

In general, polyclonal antibodies work better when bound to affinity chromatography resins for use in protein purification. They are also better as capture reagents in assays such as sandwich ELISAs. Because polyclonal antibodies are mixtures of antibodies to multiple epitopes, many antibodies bind to the target protein amplifying the signal in assays like immunohistochemistry, immunocytochemistry, immunoprecipitation, and immunofluorescence.

Monoclonal antibodies are a more uniform reagent than polyclonal antibodies because only one antibody produced, which is reactive to a single epitope on a target protein. This gives them greater specificity than a polyclonal, allowing for pairs of antibodies to be used to bind to the same target protein simultaneously, making complex experiments possible. For instance, a given cell-surface receptor can be tagged to indicate its presence with one monoclonal antibody and then have its binding site blocked by another to abrogate its function. Monoclonal antibodies have made multiplexing experiments feasible. They can be made into bispecific reagents useful for bringing different cell types together, like cytotoxic T cells and cancer cells. The specificity of monoclonal antibodies is useful for targeting particular cell types or pathways, making them of interest to pharmaceutical companies as drug products (i.e., biotherapeutics).

Which Species/Strain?

A range of vertebrate species can be used to produce polyclonal antisera. The most commonly used laboratory animals are rabbits, mice, rats, hamsters, and guinea pigs (Table 3). Typically, a single sample bleed from these animals will yield 25 mL of serum from a rabbit, 100–200 µL from a mouse, and 1–2 mL from a rat, hamster, or guinea pig. Only when large volumes of sera are required are larger animals needed. Pigs, horses, goats, sheep, and donkeys are all used commercially and can be used if large volumes of immune serum are needed.

For practical reasons, rabbits represent a good choice for the routine production of polyclonal sera. They are easy to keep and handle and can be bled safely and repeatedly, and the antibodies they produce are well characterized and easily purified. With careful management, at least 500 mL of serum can be obtained from one rabbit through the course of an immunization regime. Most laboratory rabbits are outbred, and relatively little is known of the genetics of the immune response in this species. This gives them a wider range of major histocompatibility complex (MHC) class II proteins and other immune response proteins than inbred animals.

For most purposes, smaller animals such as rats, mice, hamsters, and guinea pigs are not the animal of choice for polyclonal antibody production, mainly because only small volumes of serum can be obtained. This problem can be reduced by inducing the formation of ascites in mice, which can provide up to 10 mL of ascites fluid from a single animal, and antibody titers in ascites fluids are almost as high as serum titers. However, ascites production must be justified before most Institutional Animal Care and Use committees will allow it because of the discomfort and pain it causes to the animals.

Small rodents can be used for polyclonal antibody production when only small amounts of antibody are needed or when only tiny amounts of antigen are available. Small rodents generally respond better to lower antigen doses than larger animals. For monoclonal antibody production, both mice and rats can be used. Most laboratories use inbred strains of mice and rats. Here, the genetics of the immune response have been intensively studied. Genes that affect both the quality and quantity of the antibody response occur at many distinct loci, but those associated with the MHC class II proteins are often the most important (Allen et al. 1985). If a weak response to an antigen is seen, changing the strain of animal that is immunized can help overcome the low production. Table 4 lists the MHC class II alleles of many of the standard strains of mice. A reasonable approach for dealing with the MHC class II genetic background is to immunize one of the standard strains kept in your animal facility, and if there is no response or only a weak response, then consider alternative strains. In particular, F₁ hybrid mice and/or laboratory “outbred strains” such as Swiss Webster mice are often useful.

All immune systems display the property of self-tolerance. This actively acquired state protects the animal from autoimmune damage and can have a dominant role in determining the immunogenicity of a given antigen. Thus, at its simplest, mouse serum albumin is not immunogenic in mice but is in rabbits. In practice, immunizations should be performed in animals that are as far evolutionarily from the source of the antigen as possible. For weakly immunogenic, highly conserved proteins isolated from mammalian sources, chickens or other fowl provide a valuable alternative because their self-tolerance profile to such molecules is considerably different from that of mammals. If mice need to be used for immunizations with well-conserved antigens, NZB mice often are a good choice because they tend to make autoantibodies more readily than other strains. Knockout mice in which the gene for a specific protein (antigen) has been removed before birth offer a good alternative for high-homology target antigens. Because the animal does not express a given protein, the immune system should see it as a foreign protein and respond accordingly.

Animals from various species reach immune maturation at different times post-birth. Mice are commonly considered to be mature at 6 wk, rats at 6–8 wk, and rabbits 12 wk post-birth. It is ill advised to immunize animals before their immune systems have fully matured because tolerance can result in immunological unresponsiveness to the immunogen.

How Many Animals?

Even in genetically identical animals, a single preparation of antigen often elicits a different antibody response. When outbred animals such as rabbits are used, these differences are heightened. If the amount of antigen is not limiting, several animals should be used for any immunization scheme, and individual animals should always be screened separately. For rabbits, two animals should be used as a minimum, with three to four being preferable. For mice or other rodents, three to six is advisable.

Freund's Adjuvant

The most commonly used adjuvant for research work is Freund's adjuvant (see Protocol: Preparing and Using Adjuvants [Greenfield 2018b]). Freund's adjuvant is a water-in-oil emulsion prepared with nonmetabolizable oils. If the mixture contains killed M. tuberculosis, it is referred to as Complete Freund's adjuvant (CFA); without the bacteria, it is Incomplete Freund's adjuvant (IFA). Freund's is one of the best adjuvants for stimulating strong and prolonged responses because of the nonspecific immunopotentiation of macrophages by the surfactant and the mycobacterium. The principal disadvantage of Freund's adjuvant is that it can invoke a strong inflammatory response as well as very aggressive and persistent granulomas. For this reason, it is not used for injections of humans, and the possible side effects should be monitored carefully during antibody production in animals. To avoid the majority of the side effects, the primary injection should be given in CFA, but all boosts should be performed using IFA. The Freund's/antigen emulsion should be kept sterile to avoid Pasteurella bacterial infections, which can induce carbuncles at the injection site that could rupture and compromise the health of the animal. Because Freund's adjuvants are potentially harmful to humans, care should be taken during preparation and injection.

Ribi Adjuvant System (RAS)

Ribi adjuvants are a mixture of oil, detergent, and immunostimulator(s) (see Protocol: Preparing and Using Adjuvants [Greenfield 2018b]). A metabolizable oil, squalene, is used to minimize inflammation. Modified bacterial products have been added that have been designed to provide immunopotentiation without excessive inflammation. Trehalose dimycolate is a mycobacterial component that acts as both an immunostimulator and a surfactant, as well as aiding the antigen to bind to the oil droplets. This makes Ribi an oil-in-water emulsion, which causes less tissue damage than a water-in-oil emulsion like Freund's adjuvant. However, because Ribi uses less oil, the depot effect is reduced as well, and more frequent booster injections will be required.

Ribi adjuvants are referred to as an adjuvant system because they offer different detoxified bacterial products as immunostimulators specific for different species. For some species, this might make Ribi a better choice than Freund's. In general, Ribi is a good choice for proteins that have some hydrophobic or amphipathic properties because the adjuvant's effectiveness depends on adsorption of the antigen to the oil droplets. This can permit a better antibody response to epitopes of the native protein compared with denatured protein.

Hunter's TiterMax Adjuvant

Hunter's TiterMax is an oil/surfactant-based adjuvant that is water-in-oil based like Freund's adjuvant (see Protocol: Preparing and Using Adjuvants [Greenfield 2018b]). Unlike Freund's, it uses squalene and a synthetic nonionic surfactant (a copolymer of polyoxyethylene and polyoxypropylene) with good protein-binding capacity. The surfactant can activate complement and bind complement components that help target the antigen to follicular dendritic cells in the lymph nodes and spleen. TiterMax Gold is the latest incarnation, which works effectively in mice and rabbits. Depending on the antigen, it can work as well as or better than Freund's. The main advantage of TiterMax is that it uses copolymer-coated microparticles to form stable emulsions with less oil. This reduces the amount of inflammation at the injection site and reduces the frequency of booster injections.

Magic Mouse Adjuvant

Magic Mouse adjuvant is an aqueous suspension that has been tailor-made to induce the rapid production of high titers of antibodies in mice. The adjuvant contains immune-stimulatory CpG DNAs, which are short oligonucleotides containing unmethylated cytosine–guanine dinucleotides within a certain base context, which can be found naturally in bacterial DNA. CpG activates the innate immune system through Toll-like Receptor 9 (TLR9) found on macrophages. TLR9 triggering activates a signaling cascade leading to the production of pro-inflammatory cytokines and activation of the innate immune system. When mixed with an antigen, CpG DNA produces high titers of antigen-specific antibodies.

Because different CpG DNA sequences activate the immune systems of different animal species, Magic Mouse adjuvant is specifically designed for immunization of mice. Unlike water-in-oil emulsions, Magic Mouse adjuvant preparation does not require sonication, heating, lyophilization, or homogenization, allowing the native conformation of the immunogen to be maintained. Because Magic Mouse is aqueous based, there is no depot effect and frequent booster injections could be required. Magic Mouse is nontoxic with no adverse effects. This can be useful for generating antibodies against conformational epitopes on native antigens.

Pam3Cys-Ser-(Lys)4 Adjuvant

Pam3Cys-Ser-(Lys)4 is a lipopeptide that has been reported to be a potent stimulator of immune responses in mice and rabbits (Bessler and Jung 1992), markedly enhancing the immune response when given as a mixture with antigens. The reported side effects have been significantly less than for Freund's adjuvant while still inducing a significant antigen-specific immune response (Kellner et al. 1992). Pam3Cys-Ser-(Lys)4 can be covalently coupled with most antigens. It is a TLR2 agonist, inducing a signaling cascade effect leading to the activation of pro-inflammatory transcription factor NF-κB.

GERBU Adjuvant

GERBU adjuvant is an aqueous-phase adjuvant. This means that it lacks any depot effect, making it necessary to boost the animals frequently to obtain a high-titer response. It shows minimal inflammation at the injection sites and avoids many of the drawbacks seen with Freund's adjuvant. GERBU is mainly a colloidal suspension of cationic nanoparticles, GMDP (a glycopeptide derived from Lactobacillus bulgaricus cell walls) and cimetidine (a histamine antagonist). The nanoparticles bind the antigen and carry it to the lymphocytes, where it is phagocytized and presented to the immune system.

Aluminum Hydroxide Adjuvant

A common alternative to Freund's adjuvant is to adsorb the immunogen onto an aluminum salt (Glenny et al. 1926; see also Protocol: Preparing and Using Adjuvants [Greenfield 2018b]). Aluminum hydroxide is the most commonly used, but every aluminum salt avoids many of the more harmful side effects of Freund's adjuvants. The immunogen is either allowed to adsorb to the preformed aluminum salt or it can be trapped in the salt during precipitation. Adsorption of the protein is dependent on the pI: proteins with lower pI adsorb more strongly to the positively charged aluminum salts. When injected, the precipitate provides the depot effect.

If desired, nonspecific stimulation can be provided by adding killed bacteria, such as heat-killed B. pertussis, to the preparation. However, the addition of B. pertussis reintroduces some of the potential side effects seen with Complete Freund's adjuvant. Aluminum hydroxide adjuvants, when used without the B. pertussis stimulants, can be injected into all sites. When B. pertussis or other agents are used, intravenous injection is no longer an option.

Aluminum hydroxide adjuvants are generally much weaker than emulsion-based adjuvants. However, because of their mild inflammatory reactions, efficacy in generating memory, and fewer safety issues, they are the primary adjuvant used in humans.

Soluble Antigens

Soluble protein and peptide antigens are the easiest with which to work. They can yield strong responses and good monoclonal antibodies with doses as low as 1 µg/injection. More commonly, injections are adjusted to deliver 10–20 µg. If the antigen is available in large quantities, 50 µg should be used. Except for special cases, it is seldom worthwhile to use more than 200 µg of a protein antigen per injection. Even if the antigen is not pure, the total dose should not normally exceed 500 µg. When highly conserved proteins are being used to raise antibodies, it is often necessary to modify these antigens before injection. This can be performed by covalently adding small immunogenic haptens to proteins. Modifying proteins by binding them to large immunogenic proteins such as the hemocyanins has also been shown to be an effective way of breaking T-cell tolerance. These methods are discussed in detail in Introduction: Selecting the Antigen (Greenfield et al. 2021).

Insoluble Antigens

Particulate antigens can be much better immunogens than soluble molecules. Whole cells, bacteria, and viruses are usually very immunogenic. Soluble monomeric fractions of many proteins induce poor responses. Yet, when the same protein molecules are aggregated, they induce a good antibody response. There are also many examples of weakly antigenic proteins that have been made highly antigenic by coupling them to large matrices such as agarose beads. One of the reasons for these effects is the rapid phagocytosis of particulate antigens compared with soluble material. Most soluble antigens can be made more immunogenic by coupling them chemically to beads or cells, by converting them to larger compounds by self-polymerization (e.g., with chemical cross-linkers, or by partial denaturation), or by binding them to carrier proteins. Large, insoluble antigens should not be given intravenously because they can induce embolisms, killing the animal.

The valency of an antigen can also have a strong effect on its immunogenicity. This is particularly true for large complex carbohydrates, which are much more immunogenic than simple compounds. Apparently, the repeating epitopes on these immunogens make better targets for binding to B cells. However, polymeric antigens with nondegradable backbones can paralyze the B cell by blocking its surface receptors and preventing the immunogen from being processed.

Recombinant Proteins Produced by Overexpression

Recent advances in recombinant DNA technology have simplified the production of many protein antigens. Overexpression of fusion proteins or full-length polypeptide chains using both prokaryotic and eukaryotic vectors has become routine. These proteins are often excellent antigens and can be produced in large quantities. They normally present few problems for the production of monoclonal antibodies. These proteins can be purified and injected as soluble or insoluble antigens.

Synthetic Peptides

Synthetic peptides, when coupled to carrier proteins such as bovine serum albumin or keyhole limpet hemocyanin (KLH), elicit a strong humoral response. Constructing these carrier complexes and the production of antipeptide sera are described in Introduction: Selecting the Antigen (Greenfield et al. 2021). Using peptide–carrier protein complexes for the production of monoclonal antibodies normally is done only for specific reasons. Because these peptides are relatively short, many of the advantages of monoclonal antibody specificity are lost. Monoclonal antibodies do provide two advantages over polyclonal antipeptide sera. The first is that the source of the antibodies will be unlimited, and the second is that monoclonal antibodies can be more useful in immunoaffinity purifications. Like all immunizations using peptide antigens, the major difficulty will be in preparing antibodies that will also bind to the native protein.

Live Cells

Several studies have used live cells as immunogens for generating antibodies to surface antigens. Except in unusual circumstances, injections of cells should not include live bacteria or yeast. Although mice are normally capable of killing and clearing bacteria and yeast infections, the possibility of infecting an entire mouse colony is too great to risk these types of injections.

Although large numbers of hybridomas have been prepared to surface antigens of mammalian cells, these antibodies can be of low affinity, and care should be taken to ensure that the immune response includes antibodies that will be useful in subsequent studies. When raising antibodies to live tumorigenic cells, it is easy to pass the cells as tumors and thereby eliminate any activity against tissue culture reagents, including proteins in bovine serum.

Nucleic Acids

Nucleic acids normally are not good target antigens. Antibodies to them usually are raised against small haptens bound to carrier proteins. Because nucleic acids are weak antigens, it is particularly important to test sera for antibodies that will work in all assays for which the monoclonal antibodies are being raised.

Carbohydrates

Simple carbohydrates usually are weak immunogens. These compounds should be coupled to carrier proteins. Large complex carbohydrates (>50,000) will induce a moderate response, but often without a secondary response. High doses readily induce tolerance; thus, the injected amount should be controlled carefully. These immunogens are best injected as a portion of a larger particle, such as a bacterial cell wall or equivalent. Coupling these larger carbohydrates to carrier proteins can be beneficial. For glycoproteins, the polypeptide backbone can function as an effective carrier.

Subcutaneous Injections

Subcutaneous injections (s.c.) are used widely to immunize laboratory animals, especially rabbits (see Protocol: Routes of Antigen Injection in Rabbits [Greenfield 2018c] and Protocol: Routes of Antigen Injection in Mice and Rats [Greenfield 2019b]). Injections can include both particulate immunogens and/or adjuvants. Injected material will drain quickly into the local lymphatic system and will become concentrated in the lymph nodes closest to the injected sites. Larger animals such as rabbits normally are given multiple subcutaneous injections on the back, whereas mice are injected on the back of the neck or on one or both sides in the groin. Each injection should be relatively small, often 50–100 µL for mice, up to 400 µL for rats, and up to 800 µL for rabbits. Large injection volumes containing Freund's adjuvant should not be injected into one site because they can cause severe granulomas. The inoculum is far better delivered by multiple (up to 10 sites for a rabbit) doses in separate sites. Anesthesia is normally recommended for larger animals, but is not normally required for mice. It is used only to allow easier handling during injections.

Subcutaneous injections are used to deliver soluble or insoluble antigens into a local environment that is a good site of lymphoid draining. Maximum volumes for subcutaneous injections are about one-fifth the maximum used for intraperitoneal (i.p.) injections (100 µL compared with 500 µL). Subcutaneous injections normally are performed at more than one site to help ensure that the antigen is detected.

Like peritoneal injections (see below), subcutaneous injections that do not use adjuvants normally are used for delivering live cells to the mouse. This route is often used for tumorigenic cells.

Protein antigens immobilized on nitrocellulose often make exceptionally good immunogens. This is probably the result of their slow release from the paper, thus behaving somewhat like an adjuvant. Not all antigens show increased immunogenicity using this methodology, but some do. The antigen is bound to paper and is implanted on the back of the mouse's neck, a location that makes it difficult for the mouse to disturb the surgical clip (Protocol: Immunizing Mice and Rats with Nitrocellulose-Bound Antigen [Greenfield 2019c]).

Intramuscular Injections

Intramuscular injection (i.m.) is one of the routes of inoculation used to produce a slow release of an antigen (see Protocol: Routes of Antigen Injection in Rabbits [Greenfield 2018c] and Protocol: cDNA Immunization of Mice, Rats, and Hamsters [Greenfield 2021a]). The inoculum is deposited directly into the muscle tissue, and the antigen is seen as it drains into the nearby interstitial spaces. The inoculum will then drain into the local lymph nodes. Because of the difficulty of injecting into the muscle tissue of smaller rodents, intramuscular injections are not recommended for mice. Rabbits can be injected with volumes up to 0.5 mL, and both adjuvant and particulate inoculi can be used.

Intradermal Injections

Intradermal injection (i.d.) is used quite commonly for the immunization of larger animals (see Protocol: Routes of Antigen Injection in Rabbits [Greenfield 2018c]). Like intramuscular injections, intradermal injections are used to give a very slow rate of release of the immunogen. The inoculum is injected between the layers of the skin and is slowly absorbed into the body. Intradermal injections can be used with particulate- and adjuvant-containing inoculi.

Only small injection volumes can be used; thus, intradermal injection is only suitable if the antigen is available in a concentrated form. It is normal practice to deliver the injection over multiple sites. The procedure requires more skill than other injections. It is technically difficult to administer with small rodents and thus is seldom used for mice.

Intravenous Injections

For most laboratory animals, intravenous injection (i.v.) is a practical method of delivering a secondary or subsequent boost (see Protocol: Routes of Antigen Injection in Mice and Rats [Greenfield 2019b] and Protocol: Routes of Antigen Injection in Rabbits [Greenfield 2018c]). It is seldom used for a primary injection. Antigen delivered directly into the bloodstream will be processed rapidly by the reticuloendothelial system, primarily in the liver, lungs, and spleen. Direct intravenous injection carries serious risk to the host animal because it can readily induce pulmonary embolisms or lethal anaphylactic shock in sensitized animals. Any toxic material present in the inoculum, such as bacterial endotoxins, is particularly dangerous when delivered i.v. It is not safe or appropriate to use Freund's adjuvant with intravenous injections. Physiological buffers and salt conditions should be used if at all possible. Detergent concentration should never exceed 0.1% for ionic detergents or 0.2% for nonionic detergents.

Intravenous injections can be used for two purposes. When immunizing mice, their main use will be to deliver the final boost just before a hybridoma fusion. Intravenous injections are also useful to ensure that the antigen is seen by the immune system. A rapid and strong response can be expected because the antigen will be collected quickly in the spleen, liver, and lungs. The antigen will be processed quickly, and no continued release of the antigen into the immune system can be expected. Consequently, these types of injections produce a short-lived response.

Intraperitoneal Injections

Direct injection into the peritoneal cavity (i.p.) is particularly easy and appropriate in small rodents (see Protocol: Routes of Antigen Injection in Mice and Rats [Greenfield 2019b]). Antigens injected into the peritoneum will drain into the thoracic lymphatics and then to the vena cava. Particulate antigens or antigens in an adjuvant can both be given safely by this route. Repeated injection of Complete Freund's adjuvant by this route will induce granulomas and fibrous adhesions within the peritoneal cavity that can make subsequent disruption of the spleen difficult for hybridoma production. If the spleen is to be collected for hybridoma fusion (see Protocol: Harvesting Tissue from Immunized Mice, Rats, and Hamsters [Greenfield 2017a]), then the intraperitoneal injection should be performed on the right side. It can also cause the production of ascites fluid and even plasmacytomas in some animals. Ascites fluid induction can be an extremely useful way to obtain large quantities of polyclonal mouse antisera. It can be enhanced by a carefully timed injection of plasmacytoma cells.

Intraperitoneal injections are the most commonly used method for introducing antigens into mice. Because of the large volume of the peritoneal cavity, the volume of the immunogen can be larger for i.p. injections than for other sites. In addition, because the injections do not deliver the antigen directly into the blood system, particulate antigens can be used. Intraperitoneal injections are not often used for inoculating rabbits because of their larger size. Other routes such as subcutaneous or intramuscular are normally used.

Intraperitoneal injections can also be used to deliver live cells into the peritoneal cavity. These immunizations normally are used to introduce live mammalian cells into mice and prepare anti–cell-surface antibodies. In general, adjuvants should not be used.

One excellent method to increase the chances of an antigen being phagocytosed is to bind it to beads. Protein antigens can be bound by free amino groups to any of a number of agarose or polyacrylamide beads. The methods for coupling are discussed in Topic Introduction: Selecting the Antigen (Greenfield et al. 2021). After coupling, the beads are injected intraperitoneally with adjuvant as described in Protocol: Routes of Antigen Injection in Mice and Rats (Greenfield 2019b). These preparations should never be injected intravenously because the chance of forming embolisms is too great.

Protein antigens can be bound to nitrocellulose and injected or implanted into the peritoneum. Two approaches can be used. If the entire piece of nitrocellulose will be implanted, the subcutaneous route is suggested (Protocol: Immunizing Mice and Rats with Nitrocellulose-Bound Antigen [Greenfield 2019c]). For intraperitoneal injections, the nitrocellulose should either be fragmented or dissolved.

Injections into Lymphoid Organs

Injections directly into a lymphoid organ offer some strong theoretical advantages over other injection routes (e.g., see Sigel et al. 1983; Nilsson et al. 1987). The total dose of the immunogen is concentrated in one region that is specialized in dealing with it. The major difficulties are that the injection is technically more difficult than other injections, often requiring an operation, and many of the advantages of adjuvants are lost because they must be avoided because of their harmful side effects in the delicate lymphoid organ. As better adjuvants become available, some of these side effects might be less worrisome. However, the technical difficulties in delivering the immunogens are still severe, particularly for novices. Two compromises are to inject into the lower region of the leg or to mark draining lymph nodes with dyes to make them easier to visualize during surgery.

These are injections for the more specialized delivery of antigens. They are often appropriate for small amounts of antigen and particularly for secondary or subsequent boosts. In theory, these types of injections can be the best methods for giving a final boost, but because they demand more skill, they are not in common use. The two most frequently used sites of injection are the footpad and the spleen. In general, footpad and spleen injections should be performed only for highly specialized antigens.

The lower leg of a rabbit or rodent has some advantages for the injection of an immunogen. In general, injections into the lower region of the foot drain into the popliteal gland near the back of the knee or into the inguinal gland in the groin. If having the immunogen drain into lymph nodes is insufficient, an injection can be targeted directly to the lymph node by injecting a dye into the region between the toes. This will stain the draining lymphatics within 5–30 min (in mice) and can be used as a guide in locating the nodes (Harrell et al. 2008). After the dye injection, an operation can be performed to locate the appropriate lymph node for immunization. Injections are performed with one of the non-oil-based adjuvants or in phosphate-buffered saline alone. This procedure is tolerated better in larger animals like rats or rabbits than in mice.

Footpad injections are common sites of injections in some countries but are banned as inhumane in others. Unless there are particular reasons for performing injections into this site, we recommend using other routes. The injections should never be performed in the major weight-bearing regions of the pad. Often inflammation and swelling occur at the injection site, leading to the animal being unable to bear weight on its injected foot, sometimes resulting in lameness. In addition, boosts given in the footpad should never contain Complete Freund's adjuvant. Always use anesthetics for footpad injections.

An alternative to footpad immunizations for targeting draining lymph nodes is hock immunizations. Hock immunizations are performed in the lateral tarsal region just above the ankle. Hock immunization has all the advantages of footpad immunization for providing a strong immune response, draining into the same lymph nodes, but with significantly less impairment of mobility (see Protocol: Routes of Antigen Injection in Mice and Rats [Greenfield 2019b]).

Boosts

An increase in B cells bearing surface antibodies specific for the inoculated antigens is first detected 5–6 d after the primary injection of antigen. Antibody is usually detected in the serum from around 7 d after the injection and persists at a low level for a few days, typically reaching peak titer around Day 10. Primary responses often are very weak, particularly for readily catabolized, soluble antigens. In many cases, efficient priming can take place in the absence of detectable antibody production; thus, for practical purposes, it will not be of much value to monitor this response.

A delay is required before reintroducing the antigen into a primed animal. A minimum of 2 or 3 wk is recommended, but there is no penalty in leaving much greater intervals. Mice and rabbits will remain effectively primed for at least a year after receiving the first injection.

The response to the second injection of the same antigen is dramatically different. The number of B cells bearing antigen-specific cell-surface antibodies increases exponentially after the secondary injection, reaching a peak between Days 3 and 4. Antibodies in the serum are also detectable at this time, but peak levels are usually achieved around Days 10–14. Typically, high levels of antibody persist for ∼2–4 wk after the second injection.

The response to the third and subsequent injections broadly mirrors that of the secondary injection. Higher titers of antibody are reached, but, more importantly, the nature and quality of the antibodies present in the serum change. These changes are known as the maturation of the immune response and have considerable practical importance because they yield high-affinity antibody preparations. The intervals between secondary, tertiary, and subsequent injections can also be varied but usually need to be extended to allow the circulating level of antibody to drop enough to prevent rapid clearance of the newly injected antigen.

For rabbits, injections are normally given every 4–6 wk. One common error in immunizing rabbits is to boost every 2 wk. Although this will continually re-expose the animal to the immunogen, the effective size of the boost will be reduced by the circulating antibody. For mice, secondary and subsequent boosts should be spaced at least at 3-wk intervals.

Repetitive Immunizations at Multiple Sites (RIMMS)

Kilpatrick et al. (1997) developed an immunization protocol that capitalizes on the rapid hypermutation and affinity maturation events that occur in secondary lymphoid tissues such as lymph nodes (see Protocol: Repetitive Immunization at Multiple Sites (RIMMS) of Mice, Rats, and Hamsters [Greenfield 2020a]). Here B lymphocytes are localized and respond more quickly to antigenic challenges. The basic RIMMS immunization protocol requires that mice be immunized five times over eleven days in several subcutaneous sites proximal to draining lymph nodes. A fusion is performed on Day 13 (Fig. 1).

View larger version:

• In this window
• In a new window

Figure 1.

Comparison of the timelines for standard immunizations versus RIMMS (repetitive immunizations at multiple sites). (SQ) Subcutaneous; (i.p.) intraperitoneally; (CFA) Complete Freund's adjuvant; (IFA) Incomplete Freund's adjuvant; (PBS) phosphate-buffered saline; (LN) lymph nodes.

The RIMMS protocol relies on immunization at sites draining into lymph nodes and the use of a myeloma partner cell that had been stably transfected with Bcl-2, the founding member of the family of apoptotic regulatory proteins. Bcl-2 plays an important role in lymphoid selection processes, including the establishment of regulation of memory B cells, maintenance of self-tolerance, preventing the expansion of pathogenic B-cell populations, and rescuing cells from apoptosis. Many B cells that do not have sufficient levels or self-reactive surface immunoglobulin fail to receive a survival signal and die. Having the Bcl-2 present in the myeloma fusion partner can enhance their survival.

The RIMMS method works mainly through the innate branch of the immune system. For this reason, it is likely to respond to conformational epitopes better than classical immunizations, which use the adaptive branch of the immune system in which antigens are phagocytized by antigen-presenting cells and presented as fragments to B cells. RIMMS can also offer more epitope diversity in response to a particular antigen, although antibody affinity might be less than with classical immunizations (Table 12).

Genetic (Naked DNA) Immunization

Since first described in the early 1990s, genetic immunization has been growing in popularity (see Protocol: cDNA Immunization of Mice, Rats, and Hamsters [Greenfield 2021a]). Simply put, genetic immunization is introducing a gene in the form of cDNA directly into an animal, allowing it to be translated by the animal into protein, which can then stimulate an immune response. The protein is expressed in native conformation with posttranslational modifications that are not always achievable with recombinant protein expression, facilitating presentation of conformational epitopes and hard-to-express recombinant proteins. Genetic immunization has been successful in generating antibody responses to difficult protein targets such as G-protein–coupled receptors (GPCRs), ion channel proteins, and other multiple membrane-spanning proteins.

Genetic immunization is most effective for generating antibodies where the cDNA encodes for secreted or cell-surface proteins to make them accessible to the immune system. The route of cDNA immunization is either through the skin (binding to gold particles and using a gene gun; Tang et al. 1992; Sato et al. 1996), the quadriceps muscle (Belperron et al. 1999; Boyle and Robinson 2000), or intravenously (Bates et al. 2006). High-affinity antibodies are favored because the proteins are expressed at low levels and are constantly present for presentation to the immune system. In addition, several months can be saved obtaining sufficient amounts of properly folded, soluble recombinant protein or high-expressing transfected cells.

Genetic immunizations can be enhanced by adding immunomodulatory sequences to the cDNA to augment adjuvanticity or activate the immune system, such as those encoding for immunostimulatory sequences, cytokines (Krieg et al. 1995), or ligands that bind antigen-presenting cells (Boyle et al. 1998; Azevedo et al. 1999). It has also been reported that in vivo electroporation can improve the efficiency in which the cDNA is delivered (Chen et al. 2005).

Deciding to Boost Again or to Fuse

Three factors will influence the decision to proceed with the production of monoclonal antibodies, all related to the quality and strength of the immune response. First is whether the antibodies recognize the antigen of interest. This is the most straightforward of the factors and the simplest to determine. The second is a complicated set of properties of the antibodies themselves and the strength of the immune response. These properties are manifested as different titers of antibodies and different affinities of the antibody for the antigen. The third factor is the appearance of spurious antibody activities against unrelated antigens.

In many cases, the tests will be relatively easy and the interpretation apparent. First, the sera should be checked for antibodies that bind to the immunogen itself. For example, if a purified antigen is used, sera could easily be tested for activity in a simple antibody capture assay, or if whole cells are used, then testing for binding to the cell surface should be performed first. However, if the monoclonal antibodies will be used for tests other than these simple assays, test bleeds should be checked in assays that resemble, as closely as possible, the tests for which the antibodies are being prepared. For many antibodies, the most useful test will be immunoprecipitation of the antigen. This assay is easy when only testing a few samples, and it will identify antibodies that will be useful in a large number of tests that depend on binding to the native antigen. If, however, the antibodies will be used extensively in immunoblot analyses, in immunohistochemical staining, or in other tests in which many antibodies might fail to work, these tests should be run as well.

Second, sera should also be monitored for the concentration of specific antibodies by titering the test bleeds in the appropriate assays. As the immune response matures, higher levels of specific antibodies will be found. However, higher levels of antibodies do not necessarily mean higher affinities. If high affinity is crucial to the intended use of the antibody, the sera should be titered and compared in assays that are sensitive to antibody affinity, such as immunoprecipitation.

The third factor to consider is the appearance of antibody activities against extraneous antigens. This response could be directed against other antigens in your preparations or might be a response to other antigens in the mouse's environment, including invasion by a pathogenic organism. If the mouse is ill, do not proceed with hybridoma construction. Isolate the mouse in a separate cage and allow it to recover before continuing. If a particularly valuable antigen is being used, more care and veterinary help might be needed. If the antibody activities are to contaminating antigens in the immunogen, a decision must be made whether to proceed. In general, making monoclonal antibodies against complex and multicomponent antigens is a very useful way of isolating specific immunochemical probes, particularly when the antigen is difficult to purify further. However, if the response against the other antigens continues to increase without a concomitant strengthening of the response to the desired antigen, other approaches might need to be taken. Either other mice should be tested (other individuals or other strains) or the antigen might need to be purified further before proceeding.

Hyperimmunization

Polyclonal antibodies are often raised by hyperimmunizing animals. Hyperimmunization refers to injection schedules in which animals repeatedly are boosted with the same antigen. The resulting antibodies have many useful properties. For example, class shift sera from primary injections contain a substantial proportion of IgMs, whereas sera from hyperimmunized animals contain mostly IgG antibodies. Hyperimmune sera often have higher levels of IgG, a substantial part of which is specific antibody. Levels of 1 mg/mL antigen-specific IgG are possible. This is equivalent to 10% of the total IgG content of serum.

Affinity Maturation

The average affinity of antibodies for an antigen increases with repeated injections. This change continues through multiple rounds of immunization. Antibody affinity affects the sensitivity of many immunochemical procedures.

The basis for affinity maturation is fairly well understood in terms of clonal selection and somatic mutation. In the presence of limiting amounts of antigen, those B cells with the highest-affinity antigen receptors will compete most successfully for antigen. Thus, clones of B cells secreting high-affinity antibodies will be selected for proliferation. This process is extended by somatic mutation of variable-region genes, and under continuous selective pressure, higher-affinity antibodies will result. Repeated injection of low doses of antigen is often best to select for sera of high affinity, although these sera might not contain concentrations of antibody as high as those elicited by larger doses.

Clonal Dominance

The selection of high-affinity clones can also affect the quality of the response in other ways. A common finding when immunizing with large proteins is that the response becomes less diverse with repeated injections. A few clones secreting antibodies to a limited number of epitopes will tend to dominate the subsequent stages of hyperimmunization protocols. Because the process that selects these clones involves several random factors, variation is often seen from animal to animal. One way to avoid this problem is to immunize multiple animals and to monitor test bleeds regularly.

Test Bleeds

Except in unusual circumstances, it is seldom worthwhile to fuse antibody-secreting cells from animals that do not have a usable titer of antibodies in their serum. Periodic test bleeds collected from immunized animals should be checked for the desired antibodies (Table 13). Tests are run on small batches of serum prepared from tail bleeds of immunized mice.

The test bleed will yield small samples of polyclonal sera. These sera should be tested in assays that will detect the presence of antibodies specific for the antigen. These tests are discussed in detail in Introduction: Immunoblotting (Litovchick 2020), Introduction: Immunoprecipitation (DeCaprio and Kohl 2020), Introduction: Immunoassays (Kohl and Ascoli 2017), Introduction: Cell Staining (Rodig 2021), and Introduction: Antibody Screening Using High-Throughput Flow Cytometry (Duensing and Watson 2018).

To make appropriate comparisons in these tests, two practical matters need to be considered. First, the test bleeds should always be titered to monitor the development of the response. The appropriate dilutions will depend on the strength of the response and on the type of assay, but, in general, 1:5 or 1:10 dilutions will be satisfactory. Second, the proper negative control should be another polyclonal serum and not an unrelated monoclonal antibody. Most often, this negative control will be serum collected either from an uninjected animal or from an animal that has been boosted with an unrelated antigen. Although it is not always necessary, using serum from a test bleed collected before immunization of the animal is the best negative control. These bleeds are known as preimmune sera.

Beginning with the first test bleeds, it is essential to mark the mice so that the immune response can be monitored in individuals. There are several methods that are currently used to identify mice. These include ear punches, toe clips, and tail markings. If your animal facility has a standard method, consult them for the proper codes. If not, an acceptable method that is not harmful to the mice is to color the toes of their hindlegs with an indelible marker. This procedure is relatively easy, and, because the marks are on the back legs, the mice do not seem to work as hard to remove the markings as on other sites. Even so, the marks need to be reapplied twice a week.

If there is a large difference between the responses in individual mice, it might be worthwhile to isolate individual mice in separate cages to ensure that the proper mouse is given the final boost.

Samples of the serum should be taken 7–14 d after an injection. This timing will correspond with the peak of antibody titers for most injection routes. Normally, small samples are collected until the desired antibodies are detected and the levels have reached acceptable levels. Test bleeds normally are performed from the marginal ear vein for rabbits, from the tail vein for mice, or from the saphenous vein of rats. Retro-orbital bleeds and submandibular bleeds are other common sites for obtaining serum samples to assess antibody titers. These sites are used because they are easily accessible and do not have high numbers of nerve endings. Taking serum samples should be relatively painless for the animal, considerably less painful than giving blood is for humans. For rabbits, 5–10 mL can be collected conveniently, and this will provide more than enough serum for most tests. For mice, rats, and hamsters, 200–300 µL is the maximum amount for most test bleeds because of the smaller blood volume in these animals.

Test bleeds normally are assayed against the immunogen itself by one of the techniques described in Introduction: Immunoblotting (Litovchick 2020), Introduction: Immunoprecipitation (DeCaprio and Kohl 2020), Introduction: Immunoassays (Kohl and Ascoli 2017), Introduction: Cell Staining (Rodig 2021), or Introduction: Antibody Screening Using High-Throughput Flow Cytometry (Duensing and Watson 2018). When assessing the quality of the sera, be sure to test the antibodies in assays that resemble as closely as possible the techniques for which the antibodies are being raised. The sera should be compared with pre-bleed and other test bleeds in titrations to determine the strength of the antibody response. Comparing titrations of various test bleeds will give an accurate measure of the course of the immunization and will provide a good measure of the correct time to begin collecting large volumes of serum.

Once a good titer has developed against the antigen of interest, regular boosts and bleeds are performed to collect the maximum amount of serum. For rabbits, boosts should be spaced every 6 wk, and serum samples of 20–40 mL should be collected ∼10–12 d after each boost. For rats and hamsters, boosts should be spaced every 2–3 wk, and serum samples of 400–500 µL should be collected 10–12 d after each boost. For mice, boosts should be spaced every 2–3 wk, and serum samples of 200–300 µL should be collected 10–12 d after each boost.

Selecting an Immunization Protocol

The way that an antigen is presented to the immune system and the frequency with which it is introduced can make a significant difference between a good response and a poor one (see Protocol: Standard Immunization of Mice, Rats, and Hamsters [Greenfield 2020b], Protocol: Standard Immunization of Rabbits [Greenfield 2020c], Protocol: Repetitive Immunization at Multiple Sites (RIMMS) of Mice, Rats, and Hamsters [Greenfield 2020a], Protocol: Subtractive Immunization for Mice, Rats, and Hamsters [Greenfield 2020d], Protocol: Decoy Immunization for Mice, Rats, and Hamsters [Greenfield 2021c], Protocol: Adoptive Transfer Immunization of Mice [Greenfield 2021d], and Protocol: cDNA Immunization of Mice, Rats, and Hamsters [Greenfield 2021a]). As described above, there are many different forms that an antigen can take. A few of these forms limit the route of immunization and/or the types of screening assays that can be used. The choice of immunogen, the species of animal selected, whether to use an adjuvant, the route of immunization chosen, and the immunization schedule (timing) are all interdependent on each other. It is also important to understand the purpose the antibody is going to fulfill. Once all these factors have been examined, an immunization protocol can be established.

Deciding on the best immunization protocol is an art. Just because one specific protocol works well for one antigen does not mean that it will work on another, even if they are chemically similar. Changes in the antigen dosage or using a particular adjuvant can have either a positive or a negative influence on the immune response. Table 14 highlights some of the advantages and disadvantages of the immunization methods.

Devising an Immunization Protocol

It is always helpful to know what the intended use of the antibody will be. If the antibody will be used for western blotting, a synthetic peptide or recombinant protein could be used as an immunogen because the protein will be denatured in the western blot. If the antibody will be used to precipitate a native protein from cell lysates before running the western blot, then a correctly folded recombinant protein, transfected cell, or cDNA could be the immunogen of choice. Here, as with antibodies used in flow cytometry or immunofluorescence, the conformation of the antigen is important. When a blocking antibody is sought, conformation of the antigen will be crucial.

Next, it helps to know something about the target protein itself. This is often referred to as bioinformatics analysis. Are there any regions of the protein that are more hydrophilic? Are these areas exposed on the surface of the protein? Will an immunogenic peptide or the recombinant protein be soluble? Is the chosen peptide similar to regions in other irrelevant proteins that could lead to the antibody being nonspecific? What is the biology of the target protein? Will the antibody affect the health of the animals if they have it in their bloodstreams? The answers to these questions will help define what form the immunogen will need to be.

Once the form of the immunogen is settled on, it is important to determine if it needs to be modified before immunization. This may mean conjugating a peptide to one or more carrier proteins like KLH or ovalbumin. Will carbohydrates be important? Bacterial production systems do not make recombinant proteins with carbohydrates on them. If these are necessary for proper charge or conformation of the target protein, a baculovirus or mammalian expression system may need to be used. If the target is an integral membrane protein, the cell membrane may be significant for its structure. This might require cDNA or a transfected cell to be the form for the immunogen. Is an adjuvant necessary or helpful? Some immunogens are in buffers that can be toxic to the animals. Using an adjuvant will allow it to be slowly introduced in small amounts at a time, making it more tolerable.

The chosen species of animal needs to be addressed next. Polyclonal antibodies are usually generated in rabbits unless very large amounts of serum are required. Most monoclonal antibodies are made in mice unless the target protein is of murine origin, in which case, rats are used. If there are homology issues among human, mouse, and rat, then hamsters, guinea pigs, or rabbits are chosen. Chickens are used if the homology is so high that none of the other species responds to it.

Taking a look at the kinds of reagents available and the intended use of the final antibody will be important in deciding how the test bleeds and fusion will be screened to select the antibodies of interest and rule out the irrelevant and nonspecific hybridomas that grow out on the fusion plates. This should be a high-throughput assay that can handle the large number of samples that will be generated. It should also reflect the intended use of the antibody. You cannot assume that an ELISA-positive antibody will work by western blot or immunohistochemistry.

Finally, the immunization schedule needs to be decided on. A common immunization protocol might be as follows (Fig. 1):

• 1. For each mouse, mix 250 µL of antigen solution with 250 µL of Complete Freund's adjuvant. Inject six BALB/c female mice i.p.

• 2. After 14 d, repeat the injections, but use Incomplete Freund's adjuvant.

• 3. Collect tail bleeds from immunized mice on Day 24. Do 1:5 dilutions in PBS, and test all samples by comparison with similar dilutions of normal mouse serum in a dot blot.

• 4. On Day 35, inject all animals i.p. with Incomplete Freund's.

• 5. Day 45: Do tail bleeds and test by dot blot. All serum samples should be checked by immunoprecipitation against in vivo radiolabeled antigen preparation.

• 6. Day 56: Inject the best responder with 100 µL i.v. and 100 µL i.p. All others get intraperitoneal injection with Incomplete Freund's.

• 7. Day 59: Fuse splenocytes from the best responder.

• 8. Day 66: The first positives should be observable.

If time is critical, a more rapid immunization protocol can be selected such as RIMMS, which is also a good choice for target proteins where conformational epitopes are present. Weak immunogens could benefit from higher dosage per immunization or more frequent immunizations. If the target protein is part of a mixture, like a cell lysate, or present on the surface of transfected cells, using a subtractive immunization protocol could be helpful.

The test bleed will let you know if you have chosen well or need to re-examine all of the above. If the titer is negative or too low, you might decide to simply reboost a few times and check the titer again. If it has not improved, changing the dosage or form of the immunogen, adjuvant, or both might help. Sometimes you just have to start over with a different species or strain. Just because the titer is weak does not necessarily mean that a good hybridoma will not be obtained from the fusion—only that it is less likely. Persistence and a little luck will often lead to success.

Troubleshooting, or “What Do I Do Now?”

Here we examine some of the possible pitfalls and solutions working with antibodies can present: