Environmental assessment - CpG oligonucleotides for use as immunostimulants and adjuvants

July 16, 2018

The information in this environmental assessment was current at the time of its preparation. It is possible that the situation may have changed since that time. Please consult the Canadian Centre for Veterinary Biologics (CCVB) if you have any questions.

Table of contents


CpG oligodeoxynucleotides (ODNs) patterned after bacterial deoxyribonucleic acid (DNA) are recognized by the vertebrate innate immune system as a 'danger signal', and stimulate both humoral and cell-mediated immune responses. These DNA molecules are currently being developed for use as both vaccine adjuvants and stand-alone immunostimulatory agents. The CCVB has prepared this Environmental Assessment to evaluate the risk of allowing the use of experimental products containing these elements.

1 Introduction

1.1 Proposed action

The Canadian Center for Veterinary Biologics (CCVB), Canadian Food Inspection Agency is responsible for licensing veterinary biologics for use in Canada. The legal authority for the regulation of veterinary biologics in Canada is provided under the Health of Animals Act and Regulations. Any veterinary biologic manufactured, sold or represented for use in Canada must comply with the requirements specified by the CCVB regarding the safety, purity, potency, and efficacy of the product.

This Environmental Assessment was prepared to provide some background regarding the use and safety of CpG oligodeoxynucleotides (ODNs) as immunostimulatory agents and vaccine adjuvants. The CCVB has granted permits to facilitate further research into these molecules in Canada.

1.2 Background

The vertebrate innate immune system recognizes bacterial DNA as foreign in part through the detection of pathogen associated molecular patterns (PAMPs) by pattern recognition receptors, which are expressed by certain cells of the immune system. Unmethylated CpG dinucleotides, one type of PAMP, are frequently found in the bacterial genome; conversely, the presence of CpG is suppressed in the vertebrate genome to about one quarter of the predicted (1 per 16 dinucleotides) value, with the cytosine base being methylated in about 80% of these occurrences.

Toll-like receptors (TLR) belong to the family of pathogen recognition receptors that are triggered by PAMP expressed by pathogens. TLRs are expressed primarily on cells of immune system, but some other cells, such as barrier epithelial cells, can also express these receptors. Eleven human and 13 mouse TLR types have been identified, and TLR-9 recognizes the unmethylated CpG motifs. Once stimulated, TLR-9 initiates a response biased towards pro-inflammatory/Th1-type immunity (Krieg, 2006).

CpG ODNs have been demonstrated to enhance mucosal as well as systemic immunity, and trigger a response that promotes both humoral and cell-mediated immunity (Mutwiri 2012). Specifically, CpG ODNs have been hypothesized to directly stimulate the proliferation and differentiation of B-cells, and activate antigen presenting cells (APCs; dendritic cells, macrophages and monocytes) to increase antigen presentation and secretion of pro-inflammatory cytokines and chemokines (IL-12, INFα, TNFα, etc.). These responses, in turn, activate natural killer cells (which secrete INFγ), helper T cells (particularly Th1 cells), cytotoxic T lymphocytes, and the generation of antigen-specific antibodies. Therefore, the use of CpG motifs as immunostimulatory agents may prove to be particularly useful in stimulating protection against intracellular bacteria and some viruses.

The co-administration of CpG ODNs with either conventional or novel vaccine types has been shown to accelerate and magnify vaccine-specific immunity. However, on their own, CpG ODNs also appear to confer innate immunity against both bacteria and viruses. Consequently, the CCVB might expect to receive applications for either of these aforementioned uses.

2 Molecular and biological characteristics of parental and biological organisms

2.1 Source, description, and function of foreign genetic material

CpG-containing sequences are typically short (~20-30 nucleotides) single-stranded ODNs, which include one or more (often 3 or 4) CpG dinucleotides present in differing base contexts (CpG motifs). They have been divided into three general classes based on structure and biological function (Table 1). Interestingly, the immunoactivity of a given CpG ODN shows species-specificity, presumably due to the evolutionary genetic divergence of the pattern recognition receptors that bind CpG ODNs, and are quite variable with respect to optimal sequence and position of composite CpG motifs. In order to increase their in vivo stability, the chemical composition of the ODN is frequently modified by replacing some of the oxygen moieties with sulphur groups on the backbone, creating phosphorothioate rather than phosphodiester bonds. Since it has been shown that maintaining close proximity of the CpG motif to an antigen improves the immune response, these ODNs may be administered along with the antigen within liposomes, or even fused to the antigenic protein. Alternatively, CpG motifs are sometimes incorporated into the plasmid backbone of a DNA vaccine.

Table 1. Comparative features of CpG ODN classes
Class Structural characteristics Immunological properties
A Phosphodiester CpG motif(s)
Phosphorothioate poly-G at 5' and 3'
Forms aggregates
Strong pDC IFN-α induction
Moderate pDC maturation
Weak B cell activation
B Phosphorothioate backbone
T-rich with CpG motifs
Strong B cell activation
Strong pDC maturation
Weak pDC IFN-α induction
C Phosphorothioate backbone
5'-TCG, CpG motif in central palindrome
Forms duplexes
Good pDC IFN-α induction
Good pDC maturation
Good B cell activation

2.2 Method of accomplishing genetic modification

Typically, CpG ODNs are artificially synthesized and subsequently purified by companies specializing in the synthesis of nucleic acids.

2.3 Genetic stability, horizontal gene transfer and potential for recombination

The transfer of genetic material between organisms by means other than reproduction, also known as horizontal gene transfer, is a relatively common occurrence in microorganisms. Therefore, exposure of microorganisms to free intact CpG ODN may result in uptake and/or incorporation of the CpG ODN into the microorganism genome. However, if such an event were to occur (exposure should be limited), it is quite unlikely to confer increased infectivity or pathogenicity to the microorganism, since the DNA fragment does not possess any known function or encode a gene product. CpG sequences are found at the predicted relative abundance in most common bacteria and large viral genomes (Burge et al., 1992). The addition of conventional DNA sequences that do not encode an apparent selective advantage are a metabolic burden, and are generally not retained by a microorganism's genome.

It is feasible that injected DNA containing CpG motifs could recombine and integrate into the genome of some target animal cells; however, the likelihood of integration is somewhat reduced due to the lack of homology of CpG motifs with the vertebrate genome and the short length of CpG ODN sequences. It has been demonstrated that foreign CpG dinucleotides, upon integration into a vertebrate genome, are subject to methylation of the cytosine residue, similar to endogenous DNA (Sutter and Doerfler, 1980). Therefore, should some of these DNA elements become incorporated into the genome of target animal cells, their chemical modification should prevent self-immunoreactivity. Nonetheless, the risk of insertional mutations caused by exogenous DNA integration into gene coding regions of the genome may warrant further evaluation regarding the frequency of CpG motif genomic integration and persistence.

2.4 Host range/specificity, tissue tropism and shed/spread capabilities

CpG ODNs have been shown to induce an immune response in a variety of host ranges, including avian, piscine, porcine, bovine, equine, and primate species. When used as an adjuvant, CpG ODNs target and are internalized by certain antigen presenting cells, which is critical to their immunoactivity. However, tissue specificity of injected CpG ODNs is not limited to immune cells. For example, it has also been shown that glioblastoma cells will uptake CpG ODNs when injected into the brain as an experimental cancer therapeutic (Zhang et al., 2005). Following subcutaneous or intravenous administration in mice and rats, in addition to being localized at the administration site, an appreciable amount of CpG ODN was shown to be distributed to the liver and kidney tissue, presumably for metabolism and excretion (Palma and Cho, 2007; Noll et al., 2005).

CpG ODNs are non-infectious and non-replicating within the vaccinated animal. Studies examining the fate of labelled ODNs within mice and rats have shown that most intravenously or intraperitoneally administered phophodiester and phosphorothioate ODNs are rapidly excreted in the urine, and to a lesser extent the feces, in a degraded form of small metabolites (Geary et al., 1999; Sands et al., 1994; Cossum et al., 1994, Agrawal et al., 1991). Taken together, these findings suggest that intact CpG adjuvant would not be shed from vaccinated animals or spread to contact target or non-target animals.

2.5 Route of administration / transmission

CpG ODNs have been administered in experimental systems and clinical trials via subcutaneous, intramuscular, intraperitoneal, intravenous, intranasal, and oral routes without complication. Conversely, injection of CpG ODN into bone joints has been shown to induce inflammation and arthritis (Zeuner et al., 2003).

3 Human safety

3.1 Previous safe use

CpG ODNs are currently being evaluated in a number of clinical trials as adjuvants for vaccines targeting infectious agents and cancer (Scheiermann and Klinman 2014; Adamus and Kortylewski 2018). More than 1,000 subjects participated in the clinical trials. The dose of CpG ODN commonly used in these trials was 1.5-15 µg/kg and the schedule of administration ranged from weekly to monthly. To date, no serious adverse events have been reported that can be directly attributed to the use of CpG ODNs. The most commonly reported adverse events are headache, fatigue, muscle ache, and injection site reaction (redness, swelling, pain). However, prolonged treatment with CpG ODNs could possibly lead to autoimmune disorders (Krieg 2006; 2012).

3.2 Probability of human exposure

Human exposure to CpG adjuvants used in veterinary vaccines is likely to be limited to veterinarians, animal technicians, manufacturing staff and testing laboratory staff. Exposure to consumers through vaccinated meat animals is not expected, since CpG adjuvants have been shown to be rapidly catabolized and cleared from an inoculated animal, and veterinary biologics typically have at least a 21 day withdrawal period.

3.3 Possible outcomes of human exposure

The demonstrated safety of CpG ODNs in human clinical trials suggests that any unintended human exposure will not cause significant harm. Moreover, the hazard of eating any animal products that may contain residual CpG ODN is not expected to be great, since the vast majority of ingested DNA (endogenous and any exogenous) is efficiently degraded within the human digestive tract (Jonas et al., 2001).

3.4 Pathogenicity in humans

Although CpG motifs are modelled after bacterial DNA, these short DNA elements do not possess any infectious properties and have no chance of reverting to virulence.

3.5 Risk associated with widespread use

Since CpG ODNs are non-replicating and are predominantly degraded in vivo by the vaccinated animal, little CpG ODN will persist in the environment, so humans will have low exposure. Therefore, in light of the aforementioned reasons for low hazard, the risk associated with widespread use of the adjuvant is likewise expected to be low.

4 Animal safety

4.1 Previous safe use

CpG ODNs have been safely administered to avian (Linghua et al., 2007; 10 μg per dose), piscine (Jørgensen et al., 2003; up to 100 μg per fish), porcine (Dory et al., 2005; 200 μg per dose), ovine (Nichani et al., 2007; up to 25 mg per dose), bovine (Ioannou et al., 2002; up to 25 mg per dose), equine (Lopez et al., 2006; 250 μg per dose) and other animal species.

A series of GLP-compliant preclinical toxicity studies have been conducted with CpG ODN 1018. These single- and multiple-dose studies in rodents and nonhuman primates have collectively shown that intramuscular injection of CpG ODN 1018 in doses up to 12.5 mg/kg produced no clinically significant toxicities. Separate reproductive and genetic toxicity studies revealed no significant reproductive or mutagenic signals and there has been no induction of anti-double-stranded DNA antibodies in mice or baboons immunized with CpG ODN 1018 vaccine formulations (Higgins et al., 2007).

4.2 Fate in target and non-target species

Studies characterizing the biodistribution of ODNs following intravenous or intraperitoneal injection suggest that the exogenous DNA is rapidly cleared from the circulating blood and is localized most abundantly at the injection site, liver and kidney (Palma and Cho, 2007; Noll et al., 2005; Cossum et al., 1994; Agrawal et al., 1991). Phosphodiester and phosphorothioate ODNs appear to be extensively degraded in vivo by nucleases and their metabolites excreted in the urine and to a lesser extent the feces. Following bolus administration in mice, the tissue elimination half-lives for intact ODN from the liver and kidney were estimated at 49.5 hr and 192 hr, respectively (Geary et al., 1999).

4.3 Potential shed and/or spread from vaccinate to contact target and non-target animals

Due to the fact that CpG ODNs are non-replicating elements, highly degraded prior to excretion by vaccinated animals, it is unlikely that CpG ODNs will be shed or spread from inoculated animals.

4.4 Reversion to virulence resulting from back passage in animals

Although CpG ODNs may be modelled after bacterial DNA sequences, they do not possess any virulence. Passaging the DNA element through an animal host will not modify this property.

4.5 Effect of overdose in target and potential non-target species

Manufacturers are required to demonstrate the safety of overdose of a complete veterinary biologic. Due to the observed variability in immunoactivity of different CpG ODNs, any change in the CpG adjuvant should warrant repetition of such safety studies.

There is some evidence for CpG ODN overdose toxicity. Mice injected daily with a 60 μg dose of CpG ODN displayed systemic toxicity, including evidence of splenomegaly, lymph node hyperplasia, hepatotoxicity and even death by day 20 (Heikenwalder et al., 2004). Conversely, a separate study indicated that mice injected weekly for two months with a 50 μg dose of CpG ODN remained physically vigorous and showed no evidence of splenomegaly or hepatomegaly (Klinman et al., 1999). In addition, weekly, repeated administration of an immunostimulatory CpG ODN in monkeys at doses as high as 10 mg/kg did not cause substantial pathology (Krieg and Wagner, 2000). The overdose toxicity of CpG ODNs is more pronounced in mice due to the wider cellular distribution of TLR-9 expression in these animals when compared to primates (Ketloy et al., 2008).

Concern has been expressed about the potential for CpG adjuvants to elicit unintended auto-immune responses in target animals. Such a condition could feasibly result from the increased persistence of self-reactive lypmphocytes, since CpG ODNs tend to nonspecifically block apoptosis of activated lymphocytes. Alternatively, CpG adjuvants have been envisioned to enhance the immunogenicity of self-proteins at the site of vaccination to trigger an auto-immune response. Similar to DNA vaccines, CpG adjuvants have been suggested to produce autoimmune disease through stimulating the production of antibodies against double-stranded DNA. One study examining this latter possibility measured a 2- to 3-fold increase in anti-DNA antibodies in mice following injection of CpG ODNs; however these elevated anti-DNA antibody levels did not appear to incite autoimmune disease in the mice (Mor et al., 1997). In contrast, other data suggest that administration of CpG ODNs can increase susceptibility to autoimmune disease in certain disease models, such as lupus and arthritis (Zeuner et al., 2003). Finally, it has also been proposed that, if the CpG ODN is administered in the context of other TNFα promoting agents (such as D-galactosamine or lipopolysaccaride), an overabundance of TNFα might result in life-threatening toxic shock (Sparwasser et al., 1997). Taken together, these uncertainties indicate that further examination into the potential side-effects of using elevated or repeated doses of CpG ODNs as immunostimulatory agents is needed.

4.6 Relative safety compared to other adjuvants

CpG adjuvants do not appear to cause extensive tissue damage at the injection site. Comparison of gross injection site morphology following co-administration into mice of hepatitis B surface antigen in combination with different adjuvants demonstrated that the CpG ODN adjuvant resulted in less muscle pathology than Titermax Gold (an experimental emulsifier), Freund's complete adjuvant (a mixture of mineral oil and mycobacteria), Freund's incomplete adjuvant, and monophosphorolipid, despite producing a robust immune response. Only alum was shown to produce less (negligible) tissue damage (Weeratna et al., 2000). Similarly, a separate study indicated that tissue damage following CpG adjuvant use in rabbits was as low as that produced with alum (Ioannou et al., 2003).

Another experimental adjuvant utilized, particularly with DNA vaccines, is a cytokine and/or chemokine co-expressed (or co-administered as purified protein) with the vaccine antigen. In comparison to this type of immunostimulatory adjuvant, CpG ODNs would probably prove safer, as CpG ODNs use mimicry to trick the immune system into coordinating the expression of multiple endogenous cytokines and chemokines, rather than over-expressing one cytokine/chemokine in an unnatural context. Moreover, the cytokine/chemokine co-expression approach runs the risk of the horizontal transfer of a gene encoding a biologically active molecule.

4.7 Safety in pregnant animals and to offspring nursing vaccinated animals

A study performed in mice suggests that exposure to high dose of CpG motif (15 mg/kg) during pregnancy could result in fetal loss and morphological defects (Prater et al., 2006). However, anomalies were not observed with a 10 times lower dose. In the another study, intraperitoneal injection of rats with 20 mg/kg of CpG ODN at day 6 of pregnancy did not produce a significant decrease in the mean number of implanted fetuses or in the number of live pups delivered (Hernando-Insua et al., 2010).

Typically, vaccination is not recommended for pregnant animals. Should manufacturers wish to vaccinate pregnant or nursing animals, perhaps it would be prudent to request evidence that CpG motifs are not transferred through the placenta or mammary glands during lactation. Although studies suggest that very low levels of CpG ODN circulate in the blood after the initial injection, caution has been raised about using this type of adjuvant in pregnant or nursing animals because of fears of tolerance developing in the fetus or neonate when exposed to CpG motifs in the absence of activated APCs.

5 Affected environment

CpG ODNs are inherently non-replicative, meaning that the source of these DNA elements is limited to that synthesized by the manufacturer. Of that quantity, the majority of CpG ODN molecules would be injected into target animals, where it is extensively degraded before being excreted outside body into the environment. Therefore, the extent of release of the adjuvant to the environment is likely to be quite low.

6 Mitigative measures

6.1 Worker safety

Individuals working in the vaccine production facility, as well as veterinarians and animal technicians could be exposed to the CpG adjuvant. As was discussed in the section above on human safety, such exposure is not anticipated to be a safety concern.

6.2 Handling Vaccinated or Exposed Animals

Exposure of animal owners to CpG ODN is likely to be very low since vaccinated animals do not shed the intact DNA molecule and unintended contamination of animal hair and skin at the vaccination site is not considered to be of public health significance.

7 Conclusions and actions

Based on the available peer-reviewed literature, CpG ODNs would not be expected to cause significant harm to human health, animal health or the environment. They have the potential to improve the efficacy of both conventional and novel vaccines. As CpG ODNs show considerable variability in immunoactivity, the safety of these DNA elements needs to be evaluated on a case-by-case basis.

8 References

Adamus, T., Kortylewski, M. (2018) The revival of CpG oligonucleotide-based cancer immunotherapies. Contep. Oncol. 22(1A): 56-60.

Agrawal, S., Temsamani, J. and Tang, J.Y. (1991) Pharmacokinetics, biodistribution, and stability of oligodeoxynucleotide phosphorothioates in mice. Proc. Natl. Acad. Sci. USA 88(17): 7595-7599.

Burge, C., Campbell, A. and Karlin, S. (1992) Over- and under-representation of short oligonucleotides in DNA sequences. Proc. Natl. Acad. Sci. USA 89(4):1358-1362.

Cossum, P.A., Truong, L., Owens, S.R., Markham, P.M., Shea, J.P. and Crooke, S.T. (1994) Pharmacokinetics of a 14C-labeled phosphorothioate oligonucleotide ISIS 2105 after intradermal administration to rats. J. Pharmacol. Exp. Ther. 269(1): 89-94.

Dory, D., Béven, V., Torché, A.-M., Bougeard, S., Cariolet, R. and Jestin, A. (2005) CpG motif in ATCGAT hexamer improves DNA-vaccine efficiency against lethal Pseudorabies virus infection in pigs. Vaccine 23(36): 4532-4540.

Geary, R.S., Matson, J. and Levin, A.A. (1999) A nonradioisotope biomedical assay for intact oligonucleotide and its chain-shortened metabolites used for determination of exposure and elimination half life of antisense drugs in tissue. Anal. Biochem. 274(2): 241-248.

Heikenwalder, M., Polymenidou, M., Junt, T., Sigurdson, C., Wagner, H., Akira, S., Zinkernagel, R. and Aguzzi, A. (2004) Lymphoid follicle destruction and immunosuppression after repeated CpG oligodeoxynucleotide administration. Nat. Med. 10(2): 187-192.

Hernando-Insua, A., Rodriguez, J.M., Elias, F., Flo, J., Lopez, R., Franco, R., Lago, N., Zorzopulos, J., and Motaner, A.D. (2010) A high dose of IMT504, the PyNTTTTGT prototype immunostimulatory oligonucleotide, does not alter embryonic development in rats. Oligonucleotides 20(1): 33-35.

Higgins, D., Marshall, J.D., Taquina, P., Van Nest, G., Livingston, B.D. (2007) Immunostimulatory DNA as a vaccine adjuvant. Expert Rev. Vaccines 6(5): 747-759.

Ioannou, X.P., Griebel, P., Hecker, R., Babiuk, L.A. and van Drunen Littel-van den Hurk, S. (2002) The immunogenicity and protective efficacy of bovine herpesvirus 1 glycoprotein D plus Emulsigen are increased by formulation with CpG oligodeoxynucleotides. J. Virol. 76(18): 9002-10.

Ioannou, X.P., Gomis, S.M., Hecker, R., Babiuk, L.A. and van Drunen Littel-van den Hurk, S. (2003) Safety and efficacy of CpG-containing oligodeoxynucleotides as immunological adjuvants in rabbits. Vaccine 21(27-30): 4368-4372.

Jonas, D.A., Elmadfa, I., Engel, K.-H., Heller, K.J., Kozianowski, G., König, A., Müller, D., Narbonne, J.F., Wackernagel, W. and Kleiner, J. (2001) Safety considerations of DNA in food. Ann. Nutr. Metab. 45(6): 235-254.

Jørgensen, J.B., Johansen, L.-H., Steiro, K. and Johansen, A. (2003) CpG DNA induces protective antiviral immune responses in Atlantic salmon (Salmo salar L.). J. Virol. 77(21): 11471-11479.

Ketloy, C., Engering, A., Srichairatanakul, U., Limsalakpetch, A., Yongvanitchit, K., Pichyangkul, S., Ruxrungtham, K. (2008) Expression and function of Toll-like receptors on dendritic cells and other antigen presenting cells from non-human primates. Vet. Immunol. Immunopathol. 125(1-2): 18-30.

Klinman, D.M., Conover, J. and Coban, C. (1999) Repeated administration of synthetic oligodeoxynucleotides expressing CpG motifs provides long-term protection against bacterial infection. Infect. Immun. 67(11): 5658-5663.

Krieg, A.M. and Wagner, H. (2000) Causing a commotion in the blood: immunotherapy progresses from bacteria to bacterial DNA. Immunol. Today 21(10): 521-526.

Krieg, A.M. (2006) Therapeutic potential of Toll-like receptor activation. Nat. Rev. Drug Discov. 5: 471-484.

Krieg, A.M. (2012) CpG still rocks! Update on an accidental drug. Nucleic Acid Ther. 22: 77-89.

Linghua, Z., Xingshan, T. and Fengzhen, Z. (2007) Vaccination with Newcastle disease vaccine and CpG oligodeoxynucleotides induces specific immunity and protection against Newcastle disease virus in SPF chicken. Vaccine 115(3-4): 216-222.

Lopez., A.M., Hecker, R., Mutwiri, G., van Drunen Littel-van den Hurk, S., Babiuk, L.A. and Townsend, H.G.G. (2006) Formulation with CpG ODN enhances antibody responses to an equine influenza virus vaccine. Vet. Immunol. Immunopathol. 114(1-2): 103-110.

Mor, G., Singla, M., Steinber, A.D., Hoffman, S.L., Okuda, K. and Klinman, D.M. (1997) Do DNA vaccines induce autoimmune disease? Hum. Gene Ther. 8(3): 293-300.

Mutwiri, G. (2012) TLR9 agonists: Immune mechanisms and therapeutic potential in domestic animals. Vet. Immunol. Immunopathol. 148: 357-368.

Nichani, A.K., Arshud Dar, M., Krieg, A.M., Mirakhur, K.K., Kaushik, R.S., Griebel, P.J., Manuja, A., Townsend, H.G.G., Babiuk, L.A. and Mutwiri, G.K. (2007) Systemic innate immune responses following intrapulmonary delivery of CpG oligodeoxynucleotides in sheep. Vet. Immunol. Immunopathol. 115(3-4): 357-368.

Noll, B.O. McCluskie, M.J., Sniatala, T., Lohner, A., Yuill, S., Krieg, A.M., Schetter, C., Davis, H.L. and Uhlmann, E. (2005) Biodistribution and metabolism of immunostimulatory oligodeoxynucleotide CPG 7909 in mouse and rat tissues following subcutaneous administration. Biochem. Pharmacol. 69(6): 981-91.

Palma, E. and Cho, M.J. (2007) Improved systemic pharmacokinetics, biodistribution, and antitumor activity of CpG oligodeoxynucleotides complexed to endogenous antibodies in vivo. J. Control. Release 120(1-2): 95-103.

Prater, R.M., Johnson, V.J., Germolec, D.R., Luster, I.L. and Holladay, S.D. (2006) Maternal treatment with a high dose of CpG ODN during gestation alters fetal craniofacial and distal limb development in C57BL/6 mice. Vaccine 24: 263-271.

Sands, H., Gorey-Feret, L.J., Cocuzza, A.J., Hobbs, F.W., Chidester, D. and Trainor, G.L. (1994) Biodistribution and metabolism of internally 3H-labeled oligonucleotides. I. Comparison of a phosphodiester and a phosphorothioate. Mol. Pharmacol. 45(5): 932-943.

Scheiermann, J., Klinman D.M. (2014) Clinical evaluation of CpG oligonucleotides as adjuvants for vaccines targeting infectious diseases and cancer. Vaccine 32(48): 6377-6389.

Sparwasser, T., Miethke, T., Lipford, G., Erdmann, A., Häcker, H., Heeg, K. and Wagner, H. (1997) Macrophages sense pathogens via DNA motifs: induction of tumor necrosis factor-alpha-mediated shock. Eur. J. Immunol. 27(7): 1671-1679.

Sutter, D. and Doerfler, W. (1980) Methylation of integrated adenovirus type 12 DNA sequences in transformed cells is inversely correlated with viral gene expression. Proc. Natl. Acad. Sci. USA 77(1): 253-256.

Weeratna, R.D., McCluskie, M.J., Xu, Y. and Davis, H.L. (2000) CpG DNA induces stronger immune responses with less toxicity than other adjuvants. Vaccine 18(17): 1755-1762.

Zeuner, R.A., Verthelyi, D., Gursel, M., Ishii, K.J. and Klinman, D.M. (2003) Influence of stipulatory and suppressive DNA motifs on host susceptibility to inflammatory arthritis. Arthritis Rheum. 48(6): 1701-1707.

Zhang, Z., Weinschenk, T. and Schluesener, H.J. (2005) Uptake, cellular distribution and novel cellular binding proteins of immunostimulatory CpG oligodeoxynucleotides in glioblastoma cells. Mol. Cell. Biochem. 272(1-2): 35-46.