Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Mol Carcinog. Author manuscript; available in PMC 2010 July 1.
Published in final edited form as:
PMCID: PMC2856944

HMGB1: The Jack-of-all-Trades Protein is a Master DNA Repair Mechanic


The high mobility group protein B1 (HMGB1) is a highly abundant protein with roles in several cellular processes, including chromatin structure and transcriptional regulation, as well as an extracellular role in inflammation. HMGB1’s most thoroughly defined function is as a protein capable of binding specifically to distorted and damaged DNA, and its ability to induce further bending in the DNA once it is bound. This characteristic in part mediates its function in chromatin structure (binding to the linker region of nucleosomal DNA and increasing the instability of the nucleosome structure) as well as transcription (bending promoter DNA to enhance the interaction of transcription factors), but the functional consequences of HMGB1’s binding to damaged DNA is still an area of active investigation. In this review we describe HMGB1’s actions in the nucleotide excision repair (NER) pathway, and we discuss aspects of both the “repair shielding” and “repair enhancing” hypotheses that have been suggested. We also report information regarding HMGB1’s roles in the mismatch repair (MMR), non-homologous end-joining (NHEJ), and V(D)J recombination pathways, as well as its newly-discovered involvement in the base excision repair (BER) pathway. We further explore the potential of HMGB1 in DNA repair in the context of chromatin. The elucidation of HMGB1’s role in DNA repair is critical for the complete understanding of HMGB1’s intracellular functions, which is particularly relevant in the context of anti-HMGB1 therapies that are being developed to treat inflammatory diseases.

The HMGB1 Protein

The high mobility group B1 (HMGB1) protein (previously known as HMG1, or amphoterin), is a member of the high mobility group family of proteins. This family is separated into three groups: the HMGA (formerly HMG-I/Y) proteins, so named because they contain an A–T hook domain that binds selectively to the minor groove of AT-rich DNA; the HMGB proteins, which contain a DNA-binding B box domain that binds distorted or non-B DNA structures with high affinity and induces severe bends in the DNA; and HMGN proteins (previously named HMG-14/17), which contain a nucleosome binding domain responsible for binding to nucleosomes [1]. All of these proteins are so-called “architectural transcription factors” because they act by binding the DNA in a structure-dependent manner, and modify transcriptional regulation and chromatin structure [2]. A number of comprehensive reviews have been written about the activity of the HMG family of proteins [3, 4].

As an architectural nuclear factor, HMGB1 is capable of binding to the linker region of nucleosomal DNA [5, 6] and it competes with histone H1 to modify the dynamics of chromatin structure [7]. In addition, HMGB1 acts as a transcriptional cofactor, enhancing the association of the TBP-TATA complex with the transcriptional start site [8]. Perhaps the best demonstration of HMGB1’s critical role in transcription came in 1999, when Calogero et al. developed HMGB1 knockout mice [9], which die shortly after birth from hypoglycemia, and exhibit improper regulation of the glucocorticoid receptor. HMGB1 has also been shown to interact with and enhance the activities of a number of transcription factors implicated in cancer development, including p53 [10], retinoblastoma protein (RB) [11] and estrogen receptor (ER) [12]. These functions of HMGB1 are mediated by its ability to bind to DNA and induce further bends into the DNA.

In addition to these intracellular roles, in 1999 Wang et al. [13] demonstrated that HMGB1 is secreted from activated macrophages, and is a pathogenic mediator in the inflammatory disease sepsis. The study of HMGB1’s extracellular roles in inflammation has greatly expanded since this discovery, and HMGB1 is now being targeted for therapeutic intervention to treat sepsis and rheumatoid arthritis (reviewed in Ulloa & Mesmer [14]). In addition, when present in the extracellular matrix, HMGB1’s binding to the receptor for advanced glycation end-products (RAGE) may mediate tumor growth, invasion and metastasis (reviewed in Ellerman et al. [15]).

HMGB1 is a small (25 kDa) protein whose myriad of intracellular and extracellular roles are mediated by its relatively simple domain structure. HMGB1 contains 3 domains: the A and B box domains, which are characteristic of the HMGB family members and are responsible for binding to and bending of DNA; and a C-terminal 30 amino acid acidic tail [16]. These domains allow HMGB1 to bind DNA in a structure-specific fashion, and this ability is responsible for its intracellular roles. The remainder of this review will focus on HMGB1’s preferential binding to non-canonical DNA structures and damaged DNA, and how this affects the repair of damaged DNA. In addition, HMGB1 can be post-translationally modified, particularly acetylated, and this affects both its ability to bind and bend the DNA [17, 18], as well as its subcellular localization [19].

HMGB1 Binds to Bent DNA Structures

Not long after its identification, HMGB1 was found to bind preferentially to non-canonical DNA structures. The first substrates to be tested were single-stranded DNA [20], supercoiled DNA and Z-DNA [21] and in all cases, HMGB1 was found to bind with higher affinity to the DNA that contained altered or bent structures. Upon binding to DNA, HMGB1 carries out the second half of its characteristic function, by inducing a further distortion in the structure upon insertion of a hydrophobic wedge into the minor groove [16]. This bend that HMGB1 induces in the DNA (measured as ~77° on linear DNA [22]), allows HMGB1 to affect the structure and stability of the DNA-protein complexes in the vicinity of its binding site. When HMGB1 binds to the linker region of nucleosomal DNA [5, 6], its bending is thought to destabilize the nucleosome structure [23] (in direct contrast to histone H1, which stabilizes the nucleosome structure), and increase the mobility of chromatin. When HMGB1 binds to promoter regions, the bends it induces have been postulated to bring distant transcription factors into close vicinity, which would allow them to interact, thus enhancing transcription [10]. Over the years, HMGB1’s affinity for a number of different DNA structures has been measured (Table 1), and in addition to supercoiled, single-stranded, and Z-DNA, it was found to bind preferentially to DNA mini-circles [24], four-way junctions [2530], looped structures [31], hemi-catenated DNA [32, 33], and we have demonstrated binding to triplex DNA [34].

Table 1
HMGB1 Binding to Damaged DNA and Non-B DNA Structures

The functional consequences of this binding (and presumably bending) of distorted DNA structures were not explored, however, until it was found that HMGB1 also binds preferentially to damaged DNA. In 1992, Hughes et al. [35] identified HMGB1 in a screen for proteins that bind to cisplatinated-DNA, and subsequently a number of groups have also described this binding [3644], and have further demonstrated that HMGB1 binds specifically to the 1,2-d(GpG) or 1,2-d(ApG) cisplatin lesions, and not the 1,3-d(GpTpG) cisplatin lesion [36]. While cisplatin is HMGB1’s best-characterized damaged-DNA substrate, other types of DNA damage have also been identified as binding substrates (Table 1), including DNA damaged by chromium [45], ultraviolet radiation (UV; [42, 46]), which includes both pyrimidine (6–4) pyrimidone adducts (6–4PPs) and cyclobutane pyrimidine dimers (CPDs), acetyl aminofluorene (AAF), benzo[a]pyrene diol epoxide (BPDE; [42]), the nucleoside analog mercaptopurine [47], and we have shown binding to triplex-directed psoralen ICLs [34]. In 1999, a co-crystal structure was derived from the HMGB1 A box binding to a cisplatin-modified oligonucleotide [48], and this showed that HMGB1’s binding to the damaged DNA induced a severe bend, over and above the distortion induced by the adduct itself. Several groups pursued the study of the functional consequences of this interaction, and two hypotheses were proposed for its effects on the repair of damaged DNA. The binding of HMGB1 to the DNA lesion could block the access of the repair machinery (the “repair shielding” hypothesis; [49]), or HMGB1 could enhance the recognition and processing of the DNA damage, because the distortion it induces in the DNA should facilitate damage recognition (Figure 1; [50]). HMGB1 has been identified as a possible player in all four major DNA repair pathways: nucleotide excision repair (NER), mismatch repair (MMR), base excision repair (BER), and DNA double-strand break repair (DSBR). A brief description of the relevant DNA repair pathways, and HMGB1’s potential roles therein, are included below. HMGB1’s interacting proteins from each pathway, as well as the putative repair steps affected, are summarized in Table 2.

Figure 1
“Access-Repair-Restore” model of nucleotide excision repair (NER) in a chromosomal context
Table 2
Interaction of HMGB1 with DNA Repair Proteins

HMGB1 as a DNA Repair Protein

Nucleotide Excision Repair

The NER mechanism processes DNA damage that is helix-distorting (e.g. bulky DNA adducts generated from carcinogenic and/or chemotherapeutic agents), and is one of three defined DNA repair excision pathways. NER is perhaps the most versatile of all repair pathways and can process nearly every bulky adduct tested to date. Cognate NER lesions include cisplatin intrastrand crosslinks, and lesions induced by BPDE, AAF, and UV radiation. NER is also one of the pathways involved in the repair of more complex lesions such as triplex-directed ICLs. The global genome repair sub-pathway of NER (GG-NER) is initiated by recognition of the DNA lesion by the XPC-RAD23B protein complex, followed by verification of the lesion by XPA-RPA and the TFIIH complex. Upon verification, TFIIH unwinds the DNA helix around the damaged site, and the XPG and XPF-ERCC1 endonucleases cleave on either side of the damaged strand, creating a 25–30 nucleotide gap. This gap is filled by a complex containing RFC, RPA, PCNA, DNA polymerase δ or ε, and DNA ligase. The transcription-coupled repair sub-pathway (TC-NER) is similar to the GG-NER pathway, except that RNA polymerase stalling at the damaged site initiates DNA damage recognition, and there are several other proteins involved, including the CSA and CSB proteins. Deficiencies in the NER pathway lead to the disease xeroderma pigmentosum (XP), which confers an up to 2000-fold increased risk for the development of non-melanoma skin cancer. For a thorough review of the NER pathway, refer to Gillet & Scharer [51].

Given the affinity of HMGB1 for a number of NER substrates (e.g. DNA damaged by cisplatin, UV, BPDE, etc.), most work has focused on HMGB1’s role in this pathway. In 1992, not long after HMGB1 was found to bind to cisplatin lesions [35], Pil & Lippard [36] demonstrated that HMGB1 only binds specifically to the 1,2-d(GpG) or 1,2-d(ApG) cisplatin lesions, not the 1,3-d(GpTpG) cisplatin lesion. When Huang et al. [52] measured the incision of these various lesions in HeLa cell extracts, they found that addition of excess recombinant mammalian HMGB1 resulted in repair inhibition of the 1,2-d(GpG) cisplatin lesion but not the 1,3-d(GpTpG) lesion. This was the first demonstration that HMGB1 could act as an inhibitory factor in NER, and is the basis of the “repair shielding” hypothesis of HMGB1 action in DNA repair. This work was followed up by similar demonstrations from other groups. Malina et al. [49] showed that Chinese hamster ovary (CHO) cell extracts supplemented with recombinant mammalian HMGB1 caused inhibition of dual incision of cisplatin-DNA adducts, as well as DNA adducts formed by two cisplatin analogs, with the degree of inhibition being proportional to HMGB1’s binding affinity for the DNA lesion. Using nucleotide incorporation as an indicator of repair synthesis, Mitkova et al. [53] also demonstrated that excess purified HMGB1, when added to human tumor cell-free extracts could inhibit the repair of cisplatin-DNA lesions, and that this depended on the presence of the C-terminal acidic tail. In fact, they found that addition of the acidic tail domain alone could inhibit DNA repair synthesis. However, this finding is not consistent with the work demonstrating a DNA damage binding-dependence, because the 30 amino acid C-terminal tail is not capable of binding to DNA [49, 53, 54]. Zamble et al. [54] also showed that addition of testis-specific HMG-domain protein (tsHMG) to a reconstituted excision repair system inhibits the repair of a 1,2-d(GpG) cisplatin lesion, but this experiment was not performed using HMGB1. It is difficult to draw conclusions about HMGB1 based on the information obtained from the tsHMG protein, because this protein has less than 22% homology with the human HMGB1 protein [55]. This data is supported by work showing that HMGB1 and RPA bind competitively to cisplatin lesions using purified proteins and purified DNA substrates in vitro, and that if HMGB1 binds to the lesion first, RPA cannot displace it [38].

Despite this body of work demonstrating that HMGB1 inhibits NER in in vitro assays, little is known about how it functions in DNA repair in a cellular context (e.g. in the presence of chromatinized DNA substrates or genomic DNA). In addition to the fact that the work has been done almost exclusively in vitro, all of the studies of HMGB1’s effects on NER were performed at HMGB1 concentrations that were much higher than physiological levels [49, 5254]. These experiments led to the conclusion that HMGB1 blocks the repair of cisplatin lesions. However, this conclusion may not be supported by other information about HMGB1’s activities. HMGB1 is a highly abundant protein that is capable of binding to DNA damage. Its affinity for damaged DNA is not as high as the affinity displayed by the NER DNA damage recognition factors (see Table 1), but given that HMGB1 is up to 10-fold more abundant than those proteins, it is likely binding to DNA damage in the cell with a reasonable frequency. If this binding was then blocking the repair of the DNA lesions, as has been suggested by several groups [49, 5254], the lack of repair could cause inhibition of cellular processes (e.g. replication, transcription) that require an intact, non-damaged, DNA structure. This situation would be unfavorable and potentially lethal for a cell, which has led us to further explore these conclusions both with purified protein and substrates in vitro, and in cell-based assays. In contrast to the work presented above, we have found that recombinant human HMGB1 can bind cooperatively with RPA to the complex triplex-directed psoralen ICL lesion [34]. This is a very different lesion than the cisplatin lesion that was used previously, and so may explain the differences in behavior of HMGB1. In addition, we have demonstrated that HMGB1 can also interact with XPC and XPA proteins [56], and that knockout mouse embryonic fibroblast (MEF) cell lines have a decreased ability to remove UV irradiation-induced CPDs and 6–4 PPs from the genome, when compared to wild-type MEF cells [50]. This is consistent with our hypothesis that this DNA damage binding protein acts to facilitate DNA lesion recognition. Other groups have shown no effect of HMGB1 on the repair of UVC lesions from genomic DNA using antibodies against cyclobutane pyrimidine dimers in the same HMGB1 knockout and wild-type MEFs [57], but this may be due to a lack of sensitivity of the assay used.

Some work has also been performed to determine the effects of HMGB1 on various outcomes of DNA damage and repair, such as cell survival in response to DNA damaging agents, as well as mutagenesis. In contrast to the findings that HMGB1 inhibited repair of cisplatin lesions, Wei et al. [58] demonstrated that HMGB1 knockout MEF cells had a similar sensitivity as wild-type cells to cisplatin, using a short-term cell viability assay. On the other hand, Krynetskaia et al. [59] showed that the same HMGB1 knockout cell lines were less sensitive to the nucleoside analogs 5-fluorouracil, araC, mercaptopurine, and thiopurine than the isogenic wild-type cell lines, using the same types of cell viability assays. In contrast, when exploring the sensitivity of HMGB1 knockout cells to UV irradiation and psoralen + UVA treatment (to induce ICLs in the DNA) using long-term clonogenic assays, we found that HMGB1 knockout MEFs were more sensitive to these DNA damaging agents [50]. This discrepancy may be explained by the types of assays used, and the doses of the damaging agents used in these studies. For short-term cell viability assays, large doses of DNA damaging agents can cause cell apoptosis and necrosis. These processes are known to cause HMGB1 release from the cell, and may also involve HMGB1 in other ways [60, 61]. Therefore, HMGB1 knockout cells may be more resistant to apoptotic cell death because it is not available to be released from cells and stimulate apoptosis. In contrast, long-term clonogenic assays measure replicative cell death, where the cell does not undergo apoptosis or necrosis, but does sustain enough DNA damage that it can no longer replicate. This type of assay is therefore more likely to measure a response to DNA damage or DNA repair than one that induces direct cell death.

Only a single publication has addressed the effect of HMGB1 on another downstream cellular outcome of DNA repair, mutagenesis. Using a plasmid-based SupF mutation reporter system, we have shown that HMGB1 knockout cells form twice as many mutations following treatment with UV irradiation or triplex-directed psoralen ICL damage than HMGB1 wild-type MEFs [50]. However, we observed no change in the background level of mutations, or in the types of mutations induced in these cell lines. This is consistent with a role for HMGB1 in enhancing the kinetics of the NER pathway. If the DNA damage is repaired more slowly in the absence of HMGB1, it is more likely that replication will occur before the damage is completely removed, and the cell will have to choose a more error-prone system of repair (e.g. translesion bypass).

In summary, the information available on HMGB1’s role in the NER pathway and in the downstream-responses to DNA lesions that are processed by NER is contradictory. However, these discrepancies may be explained by the different types of assays and DNA damaging agents, as well as concentrations of the HMGB1 protein that were used in these studies. A comprehensive study of HMGB1’s effects on NER using a number of types of DNA lesions in a single system (that is not complicated by extraneous processes such as apoptosis) in the absence of HMGB1, at endogenous levels of HMGB1, and using over-abundant protein, are required to conclusively determine the role of HMGB1 in the NER pathway. Such studies are currently underway in our laboratory.

Mismatch Repair and Base Excision Repair

HMGB1 has also been implicated in the activities of the two other defined excision repair pathways in mammalian cells, MMR and BER. MMR is responsible for repairing mismatched bases as well as insertion-deletion loops, and is initiated by the MSH2-MSH6 (MutSα) or MSH2–MSH3 (MutSβ) damage recognition complexes. These proteins recruit the MutL complex, which can consist of a variety of partners, including MLH1, PMS1 or PMS2, and mediates the initial DNA incision. The protein exonuclease 1 (EXO1) then creates a gap, which is filled by DNA polymerase and DNA ligase (reviewed in Iyer et al. [62]). BER is responsible for repairing base damage that does not distort the DNA helix, as well as abasic sites and single-strand breaks. This pathway is initiated by DNA glycosylases that are specific for the lesion, and they remove the damaged base. This is followed by the activity of an AP endonuclease (APE1) that removes the sugar phosphate and creates a single base gap with a dRP DNA end. This can either be immediately filled by DNA polymerase β (pol β) because it has both dRP-lyase and polymerase activity, in a process called short-patch BER, or a more complex pathway involving the flap structure-specific endonuclease 1 (FEN-1) can be initiated, which results in an ~8 nucleotide patch (long-patch BER) (reviewed in Baute et al. [63]).

HMGB1 was identified as a factor in each of these pathways independently, using screens for MMR or BER activities. A role for HMGB1 in MMR was identified first, when Yuan et al. [64] screened for cellular activities that could complement MMR-deficient HeLa cell extract fractions. They demonstrated that recombinant human HMGB1 is involved in the initial damage recognition/incision steps of heteroduplex repair, and is able to interact with the MMR proteins MSH2 and MLH1. This same group further determined that recombinant human HMGB1 works together with RPA to mediate EXO1-catalyzed DNA excision, and that HMGB1 can replace RPA’s activity in a reconstituted human MMR system [65]. The involvement of HMGB1 in BER is the most recently defined repair role, and was described by Prasad et al. [66]. They identified HMGB1 in a screen for proteins with a dRP lyase activity (similar to DNA pol β). The authors demonstrated that purified human HMGB1 can bind to dRP lyase substrates (an intermediate in the BER pathway), and although HMGB1 also has dRP lyase activity, it is ~600-fold less active than pol β. HMGB1 immunoprecipitates with the BER proteins APE, FEN-1 and pol β, and can enhance in vitro BER under conditions where APE is limiting. HMGB1 also co-localizes with sites of base damage in intact HeLa cells. In addition, the authors showed that HMGB1 knockout MEFs were more resistant to methyl methanesulfonate than wild-type MEFs, and they conjectured that this is because HMGB1 can enhance the formation of a toxic repair intermediate.

The study of HMGB1 in MMR and BER is restricted to these three demonstrations of its involvement, and a single demonstration of a change in cytotoxicity in HMGB1 knockout cells. Further work is required to determine HMGB1’s role in these pathways, particularly in the context of an intact cell.

Double Strand Break Repair

HMGB1 has been implicated in two sub-pathways of DSBR: non-homologous end-joining (NHEJ), which is a method of ligating DNA ends that can be error-prone, and is the predominant pathway of DSBR in mammalian cells [67]; and V(D)J recombination, which is a particular breakage-rejoining pathway mediated by recombination-activating genes (RAG) 1 and 2 that is responsible for creating T-cell receptor and immunoglobulin diversity in B and T cells [68]. NHEJ is initiated by the Ku70/Ku80 proteins, which bind to the ends of double-strand breaks (DSBs), and act as a scaffold for the catalytic subunit of DNA protein kinase (DNA-PKCS), which holds the ends of the DSB together. Then nucleases or polymerases process the break to produce compatible DNA ends, which are ligated together by the DNA ligase IV/XRCC4 complex (reviewed in Weterings et al. [67]). In 1998, Yumoto et al. [69] demonstrated that purified mammalian HMGB1 could stimulate the activity of human DNA-PK in vitro, and could target it to the ends of DSBs, similar to the Ku proteins (although more weakly). In addition, it has been shown that recombinant mammalian HMGB1 can enhance the in vitro DNA ligase activity of T4 DNA ligase, by bringing the DNA ends into close proximity, while still maintaining access to the ligase [70]. This interaction has been further defined to show that purified mammalian HMGB1 acts preferentially on intramolecular DNA ends, in a distinct manner from the similar activity of histone H1 [71].

V(D)J recombination is a pathway that is closely related to NHEJ, but is initiated by deliberate DSBs at specific and highly regulated areas of the genome in B and T cells. V(D)J recombination is instigated by the activity of RAG1 and 2, cleaving the DNA at specific recombination signal sequences (RSS). These sequences are separated by linker DNA regions of 12 or 23 base pairs (12-RSS or 23-RSS, respectively), and the RAG recombinases target areas with different spacer lengths (the 12/23 rule) for functional exon assembly. The DSB ends that are produced fold into hairpin structures, which are thought to be cleaved by the Artemis nuclease. Artemis forms a complex with DNA-PK, and the XRCC4/ligase IV complex is required to ligate the DSB ends [68]. In 1997 van Gent et al. [72] and Sawchuk et al. [73] found that purified mammalian HMGB1 can stimulate the activity of RAG proteins on 23-RSS substrates, as well as on combined 12 and 23-RSS substrates. The authors also noted that with high concentrations of HMGB1, the activity of the RAG proteins on the 12 and 23-RSS substrates was inhibited. Upon further exploration of this activity, it was found that in in vitro systems, only the three recombinant mammalian proteins (RAG1 and 2, and HMGB1) were required for the initial nicking and hairpin formation, and that they did so in accordance with the 12/23 rule [74]. Swanson [75] demonstrated that recombinant mammalian HMGB1 could bind to the 12/23 substrates in vitro, and that it could do so in a complex with the RAG1 and 2 proteins, supporting the previous findings. Recombinant HMGB1 is also involved in the ability of the Ku proteins and DNA-PK to regulate the activity of the RAG proteins, preventing 12/12 and 23/23 cleavage, and enhancing the cleavage of the 12/23 substrate [76]. Once cleavage of the substrate has taken place, the RAG and HMGB1 proteins remain bound to the DNA, suggesting a role in the subsequent ligation steps of the reaction [77]. This is supported by the finding that HMGB1 preferentially enhances the activities of DNA ligase IV on intramolecular DSBs, which is the type of break found in V(D)J recombination substrates [71]. Therefore, HMGB1 appears to be involved (at least in vitro) in the activities of NHEJ and V(D)J recombination, perhaps by enhancing the activities of the proteins on their substrates, and/or by enhancing the final ligation step.

DNA Repair in a Chromatin Context

The information presented in the previous sections describing various processes of DNA repair have predominantly been defined in in vitro systems. However, DNA repair that occurs in the cell does so in the context of chromatin, which can sterically interfere with both the formation of DNA damage (acting as a protective mechanism) as well as with the DNA repair apparatus (acting to inhibit repair). DNA repair in a chromosomal context occurs in at least three stages: i) access to the DNA damage; ii) repair of the damage; and iii) restoration of the original chromatin structure [78]. HMGB1 plays an active role in chromatin accessibility, as it is capable of facilitating the activity of the chromatin remodeling proteins ACF and CHRAC, but not ISWI [23]. HMGB1 does not form a complex with these proteins, but is thought to act by binding to the end of the nucleosome and inducing bends in the DNA, which aids in the movement of nucleosomes by the chromatin remodeling complexes [79]. HMGB1 may be playing dual roles in DNA repair in a chromatin context, acting both to enhance DNA damage recognition and processing by interacting with DNA repair proteins, and also by facilitating chromatin remodeling to allow access of the repair machinery to the DNA lesions.

Of the four major repair pathways, convincing evidence of a requirement for chromatin remodeling in conjunction with DNA repair exists only for NER and DSBR, although it is logical to assume that any DNA metabolic process in a cell should involve access to the DNA and therefore require chromatin remodeling. For both NER and DSBR pathways, repair in a chromosomal context has been broken down into the “access”, “repair” and “restore steps” [78]. The access step of NER is thought to be mediated by histone modifications (primarily acetylation, but data also exists for phosphorylation and methylation), which signals for chromatin relaxation and increased accessibility to the DNA lesions [80]. The identity of the proteins responsible for this step in mammalian cells are not well-established, although several protein candidates such as p53, ING1, ING2, HMGN1 and UV-DDB have been proposed [8185]. We have also found that cells lacking HMGB1 had a loss of UV irradiation-induced global histone acetylation, suggesting that HMGB1 may be involved in the initial “access” step of NER [50]. HMGB1 may also be involved in chromatin remodeling at the local level, because it has been shown to facilitate nucleosome sliding catalyzed by the ACF or CHRAC proteins [23]. However, how these proteins are signaling these chromatin modifications is uncertain and a large amount of work remains to be done to elucidate these steps. Post-repair restoration of the chromatin structure is also a crucial step, because it maintains the appropriate epigenetic marks in the genome and prevents epigenetic instability. This stage of the repair cycle is mediated by chromatin accessibility factor 1 (CAF-1) and anti-silencing function (ASF) proteins in mammalian cells, which tether the displaced histones to the site of repair, so they can be replaced in their original location [86]. HMGB1 is also capable of facilitating the formation of chromatin on an in vitro plasmid substrate [50], and so may also be involved in the final “restore” stage of DNA repair. A model for the potential action of HMGB1 in the chromatin remodeling-NER pathway is presented in Figure 2 [50].

Figure 2
Model of the potential effects of the HMGB1 protein on DNA repair and chromatin remodeling

Much of the work defining chromatin remodeling in response to DNA DSBs occurs in relation to the phosphorylation of the histone 2 variant, H2AX. H2AX phosphorylation (termed γ-H2AX) is involved in the formation of repair complexes on the DNA, although it does not mediate DNA repair itself [80]. Other histones marks have been identified in conjunction with either HR or NHEJ sub-pathways, but the understanding of these processes remains vague, and much work remains to be done to better understand these events. Other proteins involved in the “access” stage of DNA DSBR include ATM, Ku proteins and RAD54 [86]. Given HMGB1’s potential interactions at the sites of DNA DSBs, it may also be involved in this stage, although no work has been done to demonstrate a role for HMGB1 in DNA DSBR-related chromatin remodeling. The study of chromatin involvement in DNA repair is a rapidly expanding area of research, and much more work has to be done to define the proteins and pathways involved in DNA repair in a chromosomal context, as well as to further determine HMGB1’s roles in these processes.

Concluding Remarks

Because DNA repair pathways actively protect us from mutations that drive carcinogenesis, a better understanding of the mechanisms of DNA repair is crucial. HMGB1 is a highly abundant protein involved in many cellular processes, whose defining characteristic is the ability to bind to distorted DNA, including damaged DNA. Many studies have been performed based on this ability, some showing an inhibitory role for HMGB1 in repair, and some showing a repair-enhancing role. To move forward with an understanding of HMGB1’s functions in the cell and the ramifications of those functions, further studies on the role of HMGB1 in repair are warranted. Further, HMGB1’s role in a cellular context needs to be determined, particularly because most of the studies described in this review were performed in in vitro systems. The importance of this work relates to HMGB1’s known role as a pathogenic factor in the inflammatory diseases sepsis and rheumatoid arthritis. For this reason, treatments are being developed to remove HMGB1 from the bloodstream to ameliorate the disease. However, if HMGB1 is acting to enhance DNA repair [50, 64, 66], then such treatments may affect the capacity of the patient to repair damage to their DNA, and this in turn could lead to an increased risk for cancer. For this reason, further information about the activities of HMGB1 in DNA repair must be determined, and these in turn should be used to shape the types of anti-HMGB1 therapies used to treat these inflammatory diseases. In addition, elucidating the effects of HMGB1 overexpression in the cell could be very important, because a number of cancer types, including breast and colon cancers, overexpress HMGB1 (reviewed in Ellerman et al. [15]). If HMGB1 does inhibit DNA repair when overexpressed, then this could affect how tumors with over-abundant HMGB1 respond to DNA-damaging chemotherapeutic agents.


We thank Ms. Sarah Henninger for manuscript preparation. This work was supported by grants from the NIH - National Cancer Institute (CA097175 and CA093729 to KMV), an NIEHS Center grant ES007784, and an American Legion Auxiliary fellowship to S.S.L.


1. Bustin M. Revised nomenclature for high mobility group (HMG) chromosomal proteins. Trends Biochem Sci. 2001;26(3):152–153. [PubMed]
2. Grosschedl R, Giese K, Pagel J. HMG domain proteins: architectural elements in the assembly of nucleoprotein structures. Trends Genet. 1994;10(3):94–100. [PubMed]
3. Reeves R, Adair JE. Role of high mobility group (HMG) chromatin proteins in DNA repair. DNA Repair (Amst) 2005;4(8):926–938. [PubMed]
4. Hock R, et al. HMG chromosomal proteins in development and disease. Trends Cell Biol. 2007;17(2):72–79. [PMC free article] [PubMed]
5. Schroter H, Bode J. The binding sites for large and small high-mobility-group (HMG) proteins. Studies on HMG-nucleosome interactions in vitro. Eur J Biochem. 1982;127(2):429–436. [PubMed]
6. Nightingale K, et al. Evidence for a shared structural role for HMG1 and linker histones B4 and H1 in organizing chromatin. Embo J. 1996;15(3):548–561. [PubMed]
7. Catez F, et al. Network of dynamic interactions between histone H1 and high-mobility-group proteins in chromatin. Mol Cell Biol. 2004;24(10):4321–4328. [PMC free article] [PubMed]
8. Das D, Scovell WM. The binding interaction of HMG-1 with the TATA-binding protein/TATA complex. J Biol Chem. 2001;276(35):32597–32605. [PubMed]
9. Calogero S, et al. The lack of chromosomal protein Hmg1 does not disrupt cell growth but causes lethal hypoglycaemia in newborn mice. Nat Genet. 1999;22(3):276–280. [PubMed]
10. Jayaraman L, et al. High mobility group protein-1 (HMG-1) is a unique activator of p53. Genes Dev. 1998;12(4):462–472. [PubMed]
11. Jiao Y, Wang HC, Fan SJ. Growth suppression and radiosensitivity increase by HMGB1 in breast cancer. Acta Pharmacol Sin. 2007;28(12):1957–1967. [PubMed]
12. Melvin VS, et al. The role of the C-terminal extension (CTE) of the estrogen receptor alpha and beta DNA binding domain in DNA binding and interaction with HMGB. J Biol Chem. 2004;279(15):14763–14771. [PubMed]
13. Wang H, et al. HMG-1 as a late mediator of endotoxin lethality in mice. Science. 1999;285(5425):248–251. [PubMed]
14. Ulloa L, Messmer D. High-mobility group box 1 (HMGB1) protein: friend and foe. Cytokine Growth Factor Rev. 2006;17(3):189–201. [PubMed]
15. Ellerman JE, et al. Masquerader: high mobility group box-1 and cancer. Clin Cancer Res. 2007;13(10):2836–2848. [PubMed]
16. Thomas JO, Travers AA. HMG1 and 2, and related 'architectural' DNA-binding proteins. Trends Biochem Sci. 2001;26(3):167–174. [PubMed]
17. Pasheva E, et al. In vitro acetylation of HMGB-1 and -2 proteins by CBP: the role of the acidic tail. Biochemistry. 2004;43(10):2935–2940. [PubMed]
18. Ugrinova I, et al. In vivo acetylation of HMG1 protein enhances its binding affinity to distorted DNA structures. Biochemistry. 2001;40(48):14655–14660. [PubMed]
19. Bonaldi T, et al. Monocytic cells hyperacetylate chromatin protein HMGB1 to redirect it towards secretion. Embo J. 2003;22(20):5551–5560. [PubMed]
20. Bidney DL, Reeck GR. Purification from cultured hepatoma cells of two nonhistone chromatin proteins with preferential affinity for single-stranded DNA: apparent analogy with calf thymus HMG proteins. Biochem Biophys Res Commun. 1978;85(3):1211–1218. [PubMed]
21. Hamada H, Bustin M. Hierarchy of binding sites for chromosomal proteins HMG 1 and 2 in supercoiled deoxyribonucleic acid. Biochemistry. 1985;24(6):1428–1433. [PubMed]
22. McCauley MJ, et al. HMGB binding to DNA: single and double box motifs. J Mol Biol. 2007;374(4):993–1004. [PMC free article] [PubMed]
23. Bonaldi T, et al. The DNA chaperone HMGB1 facilitates ACF/CHRAC-dependent nucleosome sliding. Embo J. 2002;21(24):6865–6873. [PubMed]
24. Webb M, et al. Structural requirements for cooperative binding of HMG1 to DNA minicircles. J Mol Biol. 2001;309(1):79–88. [PubMed]
25. Bianchi ME. Interaction of a protein from rat liver nuclei with cruciform DNA. Embo J. 1988;7(3):843–849. [PubMed]
26. Bianchi ME, Beltrame M, Paonessa G. Specific recognition of cruciform DNA by nuclear protein HMG1. Science. 1989;243(4894 Pt 1):1056–1059. [PubMed]
27. Waga S, Mizuno S, Yoshida M. Chromosomal protein HMG1 removes the transcriptional block caused by the cruciform in supercoiled DNA. J Biol Chem. 1990;265(32):19424–19428. [PubMed]
28. Gaillard C, Strauss F. Association of poly(CA).poly(TG) DNA fragments into four-stranded complexes bound by HMG1 and 2. Science. 1994;264(5157):433–436. [PubMed]
29. JR P, et al. HMG box proteins bind to four-way DNA junctions in their open conformation. Embo J. 1998;17(3):817–826. [PubMed]
30. Webb M, Thomas JO. Structure-specific binding of the two tandem HMG boxes of HMG1 to four-way junction DNA is mediated by the A domain. J Mol Biol. 1999;294(2):373–387. [PubMed]
31. Gibb CL, et al. Effect of nuclear protein HMG1 on in vitro slippage synthesis of the tandem repeat dTG × dCA. Biochemistry. 1997;36(18):5418–5424. [PubMed]
32. Gaillard C, Strauss F. High affinity binding of proteins HMG1 and HMG2 to semicatenated DNA loops. BMC Mol Biol. 2000;1:1. [PMC free article] [PubMed]
33. Jaouen S, et al. Determinants of specific binding of HMGB1 protein to hemicatenated DNA loops. J Mol Biol. 2005;353(4):822–837. [PubMed]
34. Reddy MC, Christensen J, Vasquez KM. Interplay between Human High Mobility Group Protein 1 and Replication Protein A on Psoralen-Cross-linked DNA. Biochemistry. 2005;44(11):4188–4195. [PubMed]
35. Hughes EN, Engelsberg BN, Billings PC. Purification of nuclear proteins that bind to cisplatin-damaged DNA. Identity with high mobility group proteins 1 and 2. J Biol Chem. 1992;267(19):13520–13527. [PubMed]
36. Pil PM, Lippard SJ. Specific binding of chromosomal protein HMG1 to DNA damaged by the anticancer drug cisplatin. Science. 1992;256(5054):234–237. [PubMed]
37. Dunham SU, Lippard SJ. DNA sequence context and protein composition modulate HMG-domain protein recognition of cisplatin-modified DNA. Biochemistry. 1997;36(38):11428–11436. [PubMed]
38. Patrick SM, Turchi JJ. Human replication protein A preferentially binds cisplatin-damaged duplex DNA in vitro. Biochemistry. 1998;37(24):8808–8815. [PubMed]
39. Jamieson ER, et al. Structural and kinetic studies of a cisplatin-modified DNA icosamer binding to HMG1 domain B. J Biol Chem. 1999;274(18):12346–12354. [PubMed]
40. Jamieson ER, Lippard SJ. Stopped-flow fluorescence studies of HMG-domain protein binding to cisplatin-modified DNA. Biochemistry. 2000;39(29):8426–8438. [PubMed]
41. Kasparkova J, Farrell N, Brabec V. Sequence specificity, conformation, and recognition by HMG1 protein of major DNA interstrand cross-links of antitumor dinuclear platinum complexes. J Biol Chem. 2000;275(21):15789–15798. [PubMed]
42. Lanuszewska J, Widlak P. High mobility group 1 and 2 proteins bind preferentially to DNA that contains bulky adducts induced by benzo[a]pyrene diol epoxide and N-acetoxyacetylaminofluorene. Cancer Lett. 2000;158(1):17–25. [PubMed]
43. Imamura T, et al. Interaction with p53 enhances binding of cisplatin-modified DNA by high mobility group 1 protein. J Biol Chem. 2001;276(10):7534–7540. [PubMed]
44. Jung Y, Lippard SJ. Nature of full-length HMGB1 binding to cisplatin-modified DNA. Biochemistry. 2003;42(9):2664–2671. [PubMed]
45. Wang JF, et al. High mobility group proteins 1 and 2 recognize chromium-damaged DNA. Carcinogenesis. 1997;18(2):371–375. [PubMed]
46. Pasheva EA, Pashev IG, Favre A. Preferential binding of high mobility group 1 protein to UV-damaged DNA. Role of the COOH-terminal domain. J Biol Chem. 1998;273(38):24730–24736. [PubMed]
47. Krynetski EY, et al. A nuclear protein complex containing high mobility group proteins B1 and B2, heat shock cognate protein 70, ERp60, and glyceraldehyde-3-phosphate dehydrogenase is involved in the cytotoxic response to DNA modified by incorporation of anticancer nucleoside analogues. Cancer Res. 2003;63(1):100–106. [PubMed]
48. Ohndorf UM, et al. Basis for recognition of cisplatin-modified DNA by high-mobility-group proteins. Nature. 1999;399(6737):708–712. [PubMed]
49. Malina J, et al. Recognition of major DNA adducts of enantiomeric cisplatin analogs by HMG box proteins and nucleotide excision repair of these adducts. Chem Biol. 2002;9(5):629–638. [PubMed]
50. Lange SS, Mitchell DL, Vasquez KM. High mobility group protein B1 enhances DNA repair and chromatin modification after DNA damage. Proc Natl Acad Sci U S A. 2008;105(30):10320–10325. [PubMed]
51. Gillet LC, Scharer OD. Molecular mechanisms of mammalian global genome nucleotide excision repair. Chem Rev. 2006;106(2):253–276. [PubMed]
52. Huang JC, et al. HMG-domain proteins specifically inhibit the repair of the major DNA adduct of the anticancer drug cisplatin by human excision nuclease. Proc Natl Acad Sci U S A. 1994;91(22):10394–10398. [PubMed]
53. Mitkova E, et al. The inhibitory effect of HMGB-1 protein on the repair of cisplatin-damaged DNA is accomplished through the acidic domain. Biochemistry. 2005;44(15):5893–5898. [PubMed]
54. Zamble DB, et al. Repair of cisplatin--DNA adducts by the mammalian excision nuclease. Biochemistry. 1996;35(31):10004–10013. [PubMed]
55. Boissonneault G, Lau YF. A testis-specific gene encoding a nuclear high-mobility-group box protein located in elongating spermatids. Mol Cell Biol. 1993;13(7):4323–4330. [PMC free article] [PubMed]
56. Lange SS, Reddy MC, Vasquez KM. Human high mobility group protein B1 interacts with the XPC-RAD23B and RPA proteins on psoralen DNA interstrand crosslinks. In Preparation. 2009
57. Giavara S, et al. Yeast Nhp6A/B and mammalian Hmgb1 facilitate the maintenance of genome stability. Curr Biol. 2005;15(1):68–72. [PubMed]
58. Wei M, Burenkova O, Lippard SJ. Cisplatin sensitivity in Hmbg1−/− and Hmbg1+/+ mouse cells. J Biol Chem. 2003;278(3):1769–1773. [PubMed]
59. Krynetskaia N, et al. High mobility group protein B1 is an activator of apoptotic response to antimetabolite drugs. Mol Pharmacol. 2007 [PMC free article] [PubMed]
60. Kazama H, et al. Induction of immunological tolerance by apoptotic cells requires caspase-dependent oxidation of high-mobility group box-1 protein. Immunity. 2008;29(1):21–32. [PMC free article] [PubMed]
61. Kalinowska-Herok M, Widlak P. High mobility group proteins stimulate DNA cleavage by apoptotic endonuclease DFF40/CAD due to HMG-box interactions with DNA. Acta Biochim Pol. 2008;55(1):21–26. [PubMed]
62. Iyer RR, et al. DNA mismatch repair: functions and mechanisms. Chem Rev. 2006;106(2):302–323. [PubMed]
63. Baute J, Depicker A. Base excision repair and its role in maintaining genome stability. Crit Rev Biochem Mol Biol. 2008;43(4):239–276. [PubMed]
64. Yuan F, et al. Evidence for involvement of HMGB1 protein in human DNA mismatch repair. J Biol Chem. 2004;279(20):20935–20940. [PubMed]
65. Zhang Y, et al. Reconstitution of 5'-directed human mismatch repair in a purified system. Cell. 2005;122(5):693–705. [PubMed]
66. Prasad R, et al. HMGB1 is a cofactor in mammalian base excision repair. Mol Cell. 2007;27(5):829–841. [PMC free article] [PubMed]
67. Weterings E, Chen DJ. The endless tale of non-homologous end-joining. Cell Res. 2008;18(1):114–124. [PubMed]
68. Swanson PC. The bounty of RAGs: recombination signal complexes and reaction outcomes. Immunol Rev. 2004;200:90–114. [PubMed]
69. Yumoto Y, et al. High mobility group proteins 1 and 2 can function as DNA-binding regulatory components for DNA-dependent protein kinase in vitro. J Biochem (Tokyo) 1998;124(3):519–527. [PubMed]
70. Stros M, Cherny D, Jovin TM. HMG1 protein stimulates DNA end joining by promoting association of DNA molecules via their ends. Eur J Biochem. 2000;267(13):4088–4097. [PubMed]
71. Yamanaka S, et al. Nucleosome linker proteins HMGB1 and histone H1 differentially enhance DNA ligation reactions. Biochem Biophys Res Commun. 2002;292(1):268–273. [PubMed]
72. van Gent DC, et al. Stimulation of V(D)J cleavage by high mobility group proteins. Embo J. 1997;16(10):2665–2670. [PubMed]
73. Sawchuk DJ, et al. V(D)J recombination: modulation of RAG1 and RAG2 cleavage activity on 12/23 substrates by whole cell extract and DNA-bending proteins. J Exp Med. 1997;185(11):2025–2032. [PMC free article] [PubMed]
74. West RB, Lieber MR. The RAG-HMG1 complex enforces the 12/23 rule of V(D)J recombination specifically at the double-hairpin formation step. Mol Cell Biol. 1998;18(11):6408–6415. [PMC free article] [PubMed]
75. Swanson PC. Fine structure and activity of discrete RAG-HMG complexes on V(D)J recombination signals. Mol Cell Biol. 2002;22(5):1340–1351. [PMC free article] [PubMed]
76. Sawchuk DJ, et al. Ku70/Ku80 and DNA-dependent protein kinase catalytic subunit modulate RAG-mediated cleavage: implications for the enforcement of the 12/23 rule. J Biol Chem. 2004;279(28):29821–29831. [PubMed]
77. Hiom K, Gellert M. Assembly of a 12/23 paired signal complex: a critical control point in V(D)J recombination. Mol Cell. 1998;1(7):1011–1019. [PubMed]
78. Smerdon MJ. DNA repair and the role of chromatin structure. Curr Opin Cell Biol. 1991;3(3):422–428. [PubMed]
79. Agresti A, Bianchi ME. HMGB proteins and gene expression. Curr Opin Genet Dev. 2003;13(2):170–178. [PubMed]
80. Escargueil AE, et al. What histone code for DNA repair? Mutat Res. 2008;658(3):259–270. [PubMed]
81. Allison SJ, Milner J. Loss of p53 has site-specific effects on histone H3 modification, including serine 10 phosphorylation important for maintenance of ploidy. Cancer Res. 2003;63(20):6674–6679. [PubMed]
82. Birger Y, et al. Chromosomal protein HMGN1 enhances the rate of DNA repair in chromatin. Embo J. 2003;22(7):1665–1675. [PubMed]
83. Wang J, Chin MY, Li G. The novel tumor suppressor p33ING2 enhances nucleotide excision repair via inducement of histone H4 acetylation and chromatin relaxation. Cancer Res. 2006;66(4):1906–1911. [PubMed]
84. Kuo WH, et al. The ING1b tumor suppressor facilitates nucleotide excision repair by promoting chromatin accessibility to XPA. Exp Cell Res. 2007;313(8):1628–1638. [PubMed]
85. Wang H, et al. Histone H3 and H4 ubiquitylation by the CUL4-DDB-ROC1 ubiquitin ligase facilitates cellular response to DNA damage. Mol Cell. 2006;22(3):383–394. [PubMed]
86. Green CM, Almouzni G. When repair meets chromatin. First in series on chromatin dynamics. EMBO Rep. 2002;3(1):28–33. [PubMed]