Clinical characteristics with inflammation profiling of long COVID and association with 1-year recovery following hospitalisation in the UK: a prospective observational study

BACKGROUND

No effective pharmacological or non-pharmacological interventions exist for patients with long COVID. We aimed to describe recovery 1 year after hospital discharge for COVID-19, identify factors associated with patient-perceived recovery, and identify potential therapeutic targets by describing the underlying inflammatory profiles of the previously described recovery clusters at 5 months after hospital discharge.

METHODS

The Post-hospitalisation COVID-19 study (PHOSP-COVID) is a prospective, longitudinal cohort study recruiting adults (aged ≥18 years) discharged from hospital with COVID-19 across the UK. Recovery was assessed using patient-reported outcome measures, physical performance, and organ function at 5 months and 1 year after hospital discharge, and stratified by both patient-perceived recovery and recovery cluster. Hierarchical logistic regression modelling was performed for patient-perceived recovery at 1 year. Cluster analysis was done using the clustering large applications k-medoids approach using clinical outcomes at 5 months. Inflammatory protein profiling was analysed from plasma at the 5-month visit. This study is registered on the ISRCTN Registry, ISRCTN10980107, and recruitment is ongoing.

FINDINGS

2320 participants discharged from hospital between March 7, 2020, and April 18, 2021, were assessed at 5 months after discharge and 807 (32·7%) participants completed both the 5-month and 1-year visits. 279 (35·6%) of these 807 patients were women and 505 (64·4%) were men, with a mean age of 58·7 (SD 12·5) years, and 224 (27·8%) had received invasive mechanical ventilation (WHO class 7-9). The proportion of patients reporting full recovery was unchanged between 5 months (501 [25·5%] of 1965) and 1 year (232 [28·9%] of 804). Factors associated with being less likely to report full recovery at 1 year were female sex (odds ratio 0·68 [95% CI 0·46-0·99]), obesity (0·50 [0·34-0·74]) and invasive mechanical ventilation (0·42 [0·23-0·76]). Cluster analysis (n=1636) corroborated the previously reported four clusters: very severe, severe, moderate with cognitive impairment, and mild, relating to the severity of physical health, mental health, and cognitive impairment at 5 months. We found increased inflammatory mediators of tissue damage and repair in both the very severe and the moderate with cognitive impairment clusters compared with the mild cluster, including IL-6 concentration, which was increased in both comparisons (n=626 participants). We found a substantial deficit in median EQ-5D-5L utility index from before COVID-19 (retrospective assessment; 0·88 [IQR 0·74-1·00]), at 5 months (0·74 [0·64-0·88]) to 1 year (0·75 [0·62-0·88]), with minimal improvements across all outcome measures at 1 year after discharge in the whole cohort and within each of the four clusters.

INTERPRETATION

The sequelae of a hospital admission with COVID-19 were substantial 1 year after discharge across a range of health domains, with the minority in our cohort feeling fully recovered. Patient-perceived health-related quality of life was reduced at 1 year compared with before hospital admission. Systematic inflammation and obesity are potential treatable traits that warrant further investigation in clinical trials.

FUNDING

UK Research and Innovation and National Institute for Health Research.

Wavelength dependence of ultraviolet-induced DNA damage distribution: involvement of direct or indirect mechanisms and possible artefacts

DNA damage profiles have been established in plasmid DNA using purified DNA repair enzymes and a plasmid relaxation assay, following exposure to UVC, UVB, UVA or simulated sunlight (SSL). Cyclobutane pyrimidine dimers (CPDs) are revealed as T4 endonuclease V-sensitive sites, oxidation products at purine and pyrimidine as Fpg- and Nth-sensitive sites, and abasic sites are detected by Nfo protein from Escherichia coli. CPDs are readily detected after UVA exposure, though produced 10(3) and 10(5) times less efficiently than by UVB or UVC, respectively. We demonstrate that CPDs are induced by UVA radiation and not by contaminating UVB wavelengths. Furthermore, they are produced at doses compatible with human exposure and are likely to contribute to the mutagenic specificity of UVA [E. Sage et al., Proc. Natl. Acad. Sci. USA 93 (1996) 176-180]. Oxidative damage is induced with a linear dose dependence, for each region of the solar spectrum, with the exception of oxidized pyrimidine and abasic sites, which are not detectable after UVB irradiation. The distribution of the different classes of photolesions varies markedly, depending on wavelengths. However, the unexpectedly high yield of oxidative lesions, as compared to CPDs, by UVA and SSL led us to investigate their production mechanism. An artificial formation of hydroxyl radicals is observed, which depends on the material of the sample holder used for UVA irradiation and is specific for long UV wavelengths. Our study sheds light on a possible artefact in the production of oxidative damage by UVA radiation. Meanwhile, after eliminating some potential sources of the artefact ratios of CPDs to oxidized purine of three and five upon irradiation with UVA and SSL, respectively, are still observed, whereas these ratios are about 140 and 200 after UVC and UVB irradiation.

Chemically modified nucleic acids as immunodetectable probes in hybridization experiments

Guanine residues in nucleic acids can be modified by treatment with N-acetoxy-N-2-acetylaminofluorene and its 7-iodo derivative in an in vitro nonenzymatic reaction. The modified nucleic acids (ribo or deoxyribo, single or double stranded) are recognized by specific antibodies. They can be immunoprecipitated or used as probes in hybridization experiments and detected by immunochemical techniques.

Use of the alkaline comet assay to detect DNA repair deficiencies in human fibroblasts exposed to UVC, UVB, UVA and gamma-rays

The alkaline single cell gel electrophoresis (comet) assay applied to human fibroblasts allowed us to analyze the response to components of the solar spectrum (UVB and UVA) in comparison with the well-established response to UVC and gamma-rays. DNA strand breaks related to nucleotide excision repair of DNA photoproducts were produced 1 h after exposure to UVB or UVC in the normal cell line but not in the repair deficient XPD and TTD-2 cell lines. In contrast, the immediate production of DNA strand breaks observed in all cell lines after exposure to UVA or gamma-rays was followed by restitution of high molecular weight DNA upon post-exposure incubation. These results imply that (1) fibroblasts as well as lymphocytes can be analysed by the comet assay and (2) the comet assay clearly distinguishes cellular nucleotide excision repair capacity without the use of inhibitors of DNA synthesis.

Conformation of poly(dG-dC) . poly(dG-dC) modified by the carcinogens N-acetoxy-N-acetyl-2-aminofluorene and N-hydroxy-N-2-aminofluorene

Poly(dG-dC) . poly(dG-dC) was modified by reaction with N-acetoxy-N-acetyl-2-aminofluorene (N-AcO-AAF). Two samples with 6.6% and 8.5% modified bases were prepared. The modified bases are randomly distributed along the polymer chain, as deduced from competition experiments between antibodies against N-2-(guanosin-8-yl)-acetylaminofluorene, modified poly(dG-dC) . poly(dG-dC), and modified DNAs. Circular dichroism studies show that poly(dG-dC) . poly(dG-dC) modified by N-AcO-AAF is much more sensitive to the addition of alcohol than poly(dG-dC) . poly(dG-dC). In about 50% (vol/vol) alcohol, both polynucleotides have the same conformation, which is the Z form or a Z-like form. Moreover, in low salt and in the absence of alcohol, poly(dG-dC) . poly(dG-dC) modified by N-AcO-AAF is partially in the Z form. Poly(dG-dC) . poly(dG-dC) modified by N-hydroxy-N-2-aminofluorene can also adopt the Z form, but the transition is induced at a higher percentage than that of poly(dG-dC) . poly(dG-dC) modified by N-AcO-AAF. In low salt and in the absence of alcohol, no Z form was detected in poly(dG-dC) . poly(dG-dC) modified by N-hydroxy-N-2-aminofluorene.

Antibody-Mediated Rejection in Lung Transplantation: Clinical Outcomes and Donor-Specific Antibody Characteristics

In the context of lung transplant (LT), because of diagnostic difficulties, antibody-mediated rejection (AMR) remains a matter of debate. We retrospectively analyzed an LT cohort at Foch Hospital to demonstrate the impact of AMR on LT prognosis. AMR diagnosis requires association of clinical symptoms, donor-specific antibodies (DSAs), and C4d(+) staining and/or histological patterns consistent with AMR. Prospective categorization split patients into four groups: (i) DSA positive, AMR positive (DSA(pos) AMR(pos) ); (ii) DSA positive, AMR negative (DSA(pos) AMR(neg) ); (iii) DSA limited, AMR negative (DSA(Lim) ; equal to one specificity, with mean fluorescence intensity of 500-1000 once); and (iv) DSA negative, AMR negative (DSA(neg) ). AMR treatment consisted of a combination of plasmapheresis, intravenous immunoglobulin and rituximab. Among 206 transplanted patients, 10.7% were DSA(pos) AMR(pos) (n = 22), 40.3% were DSA(pos) AMR(neg) (n = 84), 6% were DSA(Lim) (n = 13) and 43% were DSA(neg) (n = 88). Analysis of acute cellular rejection at month 12 showed higher cumulative numbers (mean plus or minus standard deviation) in the DSA(pos) AMR(pos) group (2.1 ± 1.7) compared with DSA(pos) AMR(neg) (1 ± 1.2), DSA(Lim) (0.75 ± 1), and DSA(neg) (0.7 ± 1.23) groups. Multivariate analysis demonstrated AMR as a risk factor for chronic lung allograft dysfunction (hazard ratio [HR] 8.7) and graft loss (HR 7.56) for DSA(pos) AMR(pos) patients. Our results show a negative impact of AMR on LT clinical course and advocate for an early active diagnostic approach and evaluation of therapeutic strategies to improve prognosis.

Sequence context effects on 8-methoxypsoralen photobinding to defined DNA fragments

The photoreaction of 8-methoxypsoralen (8-MOP) with DNA fragments of defined sequence was studied. We took advantage of the blockage by bulky adducts of the 3'-5'-exonuclease activity associated with the T4 DNA polymerase. The action of the exonuclease is stopped by biadducts as well as by monoadducts. The termination products were analyzed on sequencing gels. A strong sequence specificity was observed in the DNA photobinding of 8-MOP. The exonuclease terminates its digestion near thymine residues, mainly at potentially cross-linkable sites. There is an increasing reactivity of thymine residues in the order T less than TT much less than TTT in a GC environment. For thymine residues in cross-linkable sites, the reactivity follows the order AT much less than TA approximately TAT much less than ATA less than ATAT less than ATATAA. Repeated A-T sequences are hot spots for the photochemical reaction of 8-MOP with DNA. Both monoadducts and interstrand cross-links are formed preferentially in 5'-TpA sites. Our results highlight the role of the sequence and consequently of the conformation around a potential site in the photobinding of 8-MOP to DNA.

UVA radiation is highly mutagenic in cells that are unable to repair 7,8-dihydro-8-oxoguanine in Saccharomyces cerevisiae

UVA (320-400 nm) radiation constitutes >90% of the environmentally relevant solar UV radiation, and it has been proposed to have a role in skin cancer and aging. Because of the popularity of UVA tanning beds and prolonged periods of sunbathing, the potential deleterious effect of UVA has emerged as a source of concern for public health. Although generally accepted, the impact of DNA damage on the cytotoxic, mutagenic, and carcinogenic effect of UVA radiation remains unclear. In the present study, we investigated the sensitivity of a panel of yeast mutants affected in the processing of DNA damage to the lethal and mutagenic effect of UVA radiation. The data show that none of the major DNA repair pathways, such as base excision repair, nucleotide excision repair, homologous recombination, and postreplication repair, efficiently protect yeast from the lethal action of UVA radiation. In contrast, the results show that the Ogg1 DNA glycosylase efficiently prevents UVA-induced mutagenesis, suggesting the formation of oxidized guanine residues. Furthermore, sequence analysis of UVA-induced canavanine-resistant mutations reveals a bias in favor of GC-->TA events when compared with spontaneous or H(2)O(2)-, UVC-, and gamma-ray- induced canavanine-resistant mutations in the WT strain. Taken together, our data point out a major role of oxidative DNA damage, mostly 7,8-dihydro-8-oxoguanine, in the genotoxicity of UVA radiation in the yeast Saccharomyces cerevisiae. Therefore, the capacity of skin cells to repair 7,8-dihydro-8-oxoguanine may be a key parameter in the mutagenic and carcinogenic effect of UVA radiation in humans.

Mutagenic specificity of solar UV light in nucleotide excision repair-deficient rodent cells

To investigate the role of nucleotide excision repair (NER) in the cellular processing of carcinogenic DNA photoproducts induced by defined, environmentally relevant portions of the solar wavelength spectrum, we have determined the mutagenic specificity of simulated sunlight (310-1100 nm), UVA (350-400 nm), and UVB (290-320 nm), as well as of the "nonsolar" model mutagen 254-nm UVC, at the adenine phosphoribosyltransferase (aprt) locus in NER-deficient (ERCC1) Chinese hamster ovary (CHO) cells. The frequency distributions of mutational classes induced by UVB and by simulated sunlight in repair-deficient CHO cells were virtually identical, each showing a marked increase in tandem CC-->TT transitions relative to NER-proficient cells. A striking increase in CC-->TT events was also previously documented for mutated p53 tumor-suppressor genes from nonmelanoma tumors of NER-deficient, skin cancer-prone xeroderma pigmentosum patients, compared to normal individuals. The data therefore indicate that the aprt gene in NER-deficient cultured rodent cells irradiated with artificial solar light generates the same distinctive "fingerprint" for sunlight mutagenesis as the p53 locus in NER-deficient humans exposed to natural sunlight in vivo. Moreover, in strong contrast to the situation for repair-component CHO cells, where a significant role for UVA was previously noted, the mutagenic specificity of simulated sunlight in NER-deficient CHO cells and of natural sunlight in humans afflicted with xeroderma pigmentosum can be entirely accounted for by the UVB portion of the solar wavelength spectrum.

Findings on ambulatory electrocardiographic monitoring in subjects older than 80 years

Twenty-four-hour electrocardiograms were recorded in 50 subjects (44 women, 6 men) older than 80 years without cardiovascular disease and with normal standard electrocardiographic responses. During waking and sleeping periods, the mean sinus rates were, respectively, 78 +/- 3 and 64 +/- 1 beats/min; heart rate ranged from 43 to 180 beats/min over 24 hours. Supraventricular tachycardia (SVT) was present in 28% of the subjects. Nocturnal sinus arrhythmia was only noted in 12% of the patients; it was accompanied by sinus pauses of 1.8 to 2 seconds, and 1 woman had a transient pattern compatible with atrioventricular dissociation. Supraventricular ectopic contractions (SVECs) were present in all cases. The frequency was less than 1 per hour in 25% and more than 20 per hour in 65%. Serious supraventricular tachyarrhythmias included an episode of ectopic atrial tachycardia (1 subject), a short run of atrial fibrillation (1 subject) and of flutter (1 subject), and several episodes of supraventricular tachycardia (2 subjects), all accompanied by more than 50 SVECs per hour. The number of ventricular premature contractions (VPCs) exceeded 10 per hour in 32% and were multifocal in 18%. There were couplets in 8% and a run of 6 VPCs in 1 subject (2%). In conclusion, sinus pause and atrioventricular block are unusual in people older than 80 years without apparent heart disease. In contrast, frequent SVECs and VPCs are more common. This study stresses the difficulty of evaluating the normality of the electrocardiogram with portable monitoring in the older population.

Distribution and repair of bipyrimidine photoproducts in solar UV-irradiated mammalian cells. Possible role of Dewar photoproducts in solar mutagenesis

In order to better understand the relative contribution of the different UV components of sunlight to solar mutagenesis, the distribution of the bipyrimidine photolesions, cyclobutane pyrimidine dimers (CPD), (6-4) photoproducts ((6-4)PP), and their Dewar valence photoisomers (DewarPP) was examined in Chinese hamster ovary cells irradiated with UVC, UVB, or UVA radiation or simulated sunlight. The absolute amount of each type of photoproduct was measured by using a calibrated and sensitive immuno-dot-blot assay. As already established for UVC and UVB, we report the production of CPD by UVA radiation, at a yield in accordance with the DNA absorption spectrum. At biologically relevant doses, DewarPP were more efficiently produced by simulated solar light than by UVB (ratios of DewarPP to (6-4)PP of 1:3 and 1:8, respectively), but were detected neither after UVA nor after UVC radiation. The comparative rates of formation for CPD, (6-4)PP and DewarPP are 1:0.25 for UVC, 1:0. 12:0.014 for UVB, and 1:0.18:0.06 for simulated sunlight. The repair rates of these photoproducts were also studied in nucleotide excision repair-proficient cells irradiated with UVB, UVA radiation, or simulated sunlight. Interestingly, DewarPP were eliminated slowly, inefficiently, and at the same rate as CPD. In contrast, removal of (6-4)PP photoproducts was rapid and completed 24 h after exposure. Altogether, our results indicate that, in addition to CPD and (6-4)PP, DewarPP may play a role in solar cytotoxicity and mutagenesis.

Sequence specificity in photoreaction of various psoralen derivatives with DNA: role in biological activity

The sequence specificity in the photoreaction of various psoralen derivatives with DNA is investigated by using DNA sequencing methodology. The 3'-5' exonuclease activity associated with T4 DNA polymerase serves as a probe to map the psoralens' photoaddition (monoadducts plus biadducts) on DNA fragments of defined sequence. This approach has already allowed us to demonstrate a strong sequence context effect on the 8-methoxypsoralen photobinding to DNA [Sage, E., & Moustacchi, E. (1987) Biochemistry 26, 3307-3314]. The psoralens studied include bifunctional derivatives [8-methoxypsoralen, 5-methoxypsoralen, and 4'-(hydroxymethyl)-4,5',8-trimethylpsoralen] and monofunctional derivatives (angelicin, 3-carbethoxypsoralen, and three pyridopsoralens). Maps of photochemical binding on two DNA fragments of the lacI gene of Escherichia coli are established for all the derivatives. These maps demonstrate the following general qualitative rules in the photoreaction of the furocoumarins with DNA: thymine residues in a GC environment are cold, adjacent thymines are better targets, 5'-TpA sites are strongly preferred versus 5'-ApT, and alternating (AT)n sequences are hot spots for photoaddition. Depending on the chemical structure of the derivatives and on their affinity for DNA, some minor differences in the binding spectrum are detected. A most interesting example is 3-carbethoxypsoralen, which specifically reacts with (AT)n sites. Our observations lead us to define two types of target sites: the "strong sites", which are preferential targets for all psoralen derivatives, and the "weak sites", which are targets only for derivatives having a high affinity for DNA. The frequency of DNA lesions is much higher in the former sites.(ABSTRACT TRUNCATED AT 250 WORDS)

Roflumilast inhibits the release of chemokines and TNF-α from human lung macrophages stimulated with lipopolysaccharide

BACKGROUND AND PURPOSE

Lung macrophages are critically involved in respiratory diseases. This study assessed the effects of the PDE4 inhibitor roflumilast and its active metabolite, roflumilast N-oxide on the release of a range of chemokines (CCL2, 3, 4, CXCL1, 8, 10) and of TNF-α, from human lung macrophages, stimulated with bacterial lipopolysaccharide LPS.

EXPERIMENTAL APPROACH

Lung macrophages isolated from resected human lungs were incubated with roflumilast, roflumilast N-oxide, PGE(2), the COX inhibitor indomethacin, the COX-2 inhibitor NS-398 or vehicle and stimulated with LPS (24 h). Chemokines, TNF-α, PGE(2) and 6-keto PGF(1α) were measured in culture supernatants by immunoassay. COX-2 mRNA expression was assessed with RT-qPCR. PDE activities were determined in macrophage homogenates.

KEY RESULTS

Expression of PDE4 in lung macrophages was increased after incubation with LPS. Roflumilast and roflumilast N-oxide concentration-dependently reduced the LPS-stimulated release of CCL2, CCL3, CCL4, CXCL10 and TNF-α from human lung macrophages, whereas that of CXCL1 or CXCL8 was not altered. This reduction by the PDE4 inhibitors was further accentuated by exogenous PGE(2) (10 nM) but abolished in the presence of indomethacin or NS-398. Conversely, addition of PGE(2) (10 nM), in the presence of indomethacin restored inhibition by roflumilast. LPS also increased PGE(2) and 6-keto PGF(1α) release from lung macrophages which was associated with an up-regulation of COX-2 mRNA.

CONCLUSIONS AND IMPLICATIONS

Roflumilast and roflumilast N-oxide reduced LPS-induced release of CCL2, 3, 4, CXCL10 and TNF-α in human lung macrophages.

Oxidative DNA damage photo-induced by 3-carbethoxypsoralen and other furocoumarins. Mechanisms of photo-oxidation and recognition by repair enzymes

DNA photosensitization by several furocoumarins (including 3-carbethoxypsoralen (3-CPs), 8-methoxypsoralen (8-MOP), 5-methoxypsoralen (5-MOP) and angelicin was investigated by using DNA sequencing methodology. 3-CPs induces photo-oxidation of guanine residues leading to alkali-labile sites in DNA (revealed by hot piperidine), whereas 8-MOP, 5-MOP and angelicin do not. There is a preferential photo-oxidation of G when located on the 5' side of GG doublets, likely to reflect a better accessibility of the G moiety in such a context. Mechanisms operating via both radicals (type I) and singlet oxygen (type II) are involved in the photo-oxidation of G residues by 3-CPs. Photo-oxidized G residues are produced independently of the formation of photoadducts, and scavengers of singlet oxygen or radicals do not inhibit photobinding of 3-CPs to DNA. This leads us to propose that covalent photoadducts arise from the intercalated excited sensitizer molecules, whereas G photo-oxidations are produced either by electron transfer reactions involving bound 3-CPs or by energy transfer to molecular oxygen, thereby producing singlet oxygen that subsequently reacts with guanine bases. Quantification of both types of DNA lesions indicated that in vitro photo-oxidized G residues are produced in DNA by 3-CPs plus ultraviolet light at least to the same extent as photoadducts, under our conditions. A calf thymus redoxyendonuclease, equivalent to the endonuclease III of Escherichia coli, specific for oxidative DNA damages, recognizes and cleaves DNA at sites of photo-oxidized G residues. The extent of the cleavage by this enzyme was close to that observed by hot piperidine and followed the amount of photo-oxidized G residues produced when the lifetime of excited oxygen species is modified. The redoxyendonuclease did not incise DNA treated with 8-MOP, 5-MOP or angelicin plus ultraviolet light. The exonuclease III and endonuclease IV of E. coli also involved in the repair of oxidative DNA damage, convert the replicative form I of 3-CPs-treated DNA to replicative form II. This suggests that the lesions recognized by these enzymes are apurinic-like lesions. In view of the low toxicity and mutagenicity of 3-CPs, DNA photo-oxidation products induced by the photodynamic effect of 3-CPs are likely to be efficiently taken care of by the DNA repair system(s). It is clear that 3-CPs photo-induces several classes of DNA damage, including oxidative damage.(ABSTRACT TRUNCATED AT 400 WORDS)

Hsp70--a biomarker for tumor detection and monitoring of outcome of radiation therapy in patients with squamous cell carcinoma of the head and neck

BACKGROUND

Tumor but not normal cells frequently overexpress heat shock protein 70 (Hsp70) and present it on their cell surface (mHsp70) from where it can be actively released. Therefore, membrane (mHsp70) and soluble Hsp70 (sHsp70) were investigated as potential tumor biomarkers and for monitoring the outcome of radiation therapy.

METHODS

Biopsies and blood were collected from patients with squamous cell carcinoma of the head and neck (SCCHN) at different time points (before, during therapy and in the follow-up period). Hsp70 membrane expression was determined on single cell suspensions of tumor biopsies and reference tissues by flow cytometry, sHsp70 protein and antibody levels were determined in the serum of patients and healthy donors by ELISA and NK cell markers that are related to the presence of sHsp70 were analyzed in the patient's peripheral blood lymphocytes (PBL).

RESULTS

Tumor biopsies exhibited significantly increased mHsp70 expression levels compared to the reference tissue. Soluble Hsp70 levels were significantly higher in SCCHN patients compared to healthy human volunteers and high mHsp70 expression levels on tumor cells were associated with high sHsp70 levels in the serum of patients. Following surgery and radiotherapy sHsp70 levels in patients dropped in patients without tumor relapse in the follow-up period. In contrast to sHsp70 protein, anti-Hsp70 antibody levels remained nearly unaltered in the serum of SCCHN patients before and after therapy. Furthermore, sHsp70 protein but not anti-Hsp70 antibody levels were found to be associated with the tumor volume in SCCHN patients before start of therapy. The expression densities of the activatory NK cell markers CD56, CD94, NKG2D, NKp30, Nkp44, and NKp46 differed in patients following therapeutic intervention. A significant increase in the density of NKG2D was observed in SCCHN patients in the follow-up period after surgery and radiotherapy.

CONCLUSION

We suggest sHsp70 as a potential biomarker for detecting tumors and for monitoring the clinical outcome of radiotherapy in SCCHN patients.

Formation of cyclobutane thymine dimers photosensitized by pyridopsoralens: quantitative and qualitative distribution within DNA

As after irradiation with 254-nm UV light, exposure of thymidine and three isomeric pyridopsoralen derivatives to UVA radiation, in the dry state, leads to the formation of the six diastereomers of cyclobutadithymidine as the predominant reaction. This unexpected photosensitized reaction, which also gives rise to both 5R* and 5S* diastereomers of 5,6-dihydro-5-(alpha-thymidylyl)thymidine (or "spore" photoproduct), is selective since [2 + 2] dimerization of 2'-deoxycytidine was not detected under the same experimental conditions. The cis-syn isomer of cyclobutadithymine was also found to be produced within isolated DNA following UVA irradiation in aqueous solutions containing 7-methylpyrido[3,4-c]psoralen. Quantitatively, this photoproduct represents about one-fifth of the overall yield of the furan-side pyridopsoralen [2 + 2] photocycloadducts to thymine. DNA sequencing methodology was used to demonstrate that pyridopsoralen-photosensitized DNA is a substrate for T4 endonuclease V and Escherichia coli photoreactivating enzyme, two enzymes acting specifically on cyclobutane pyrimidine dimers. Furthermore, the dimerization reaction of thymine is sequence dependent, with a different specificity from that mediated by far-UV irradiation as inferred from gel sequencing experiments. Interestingly, adjacent thymine residues are excellent targets for 7-methylpyrido[3,4-c]psoralen-mediated formation of cyclobutadithymine in TTTTA and TTAAT sites, which are also the strongest sites for photoaddition. The formation of cyclobutane thymine dimers concomitant to that of thymine-furocoumarin photoadducts and their eventual implication in the photobiological effects of the pyridopsoralens are discussed.

Cell inactivation and double-strand breaks: the role of core ionizations, as probed by ultrasoft X rays

The large RBE (approximately 7) measured for the killing of Chinese hamster V79 cells by 340 eV ultrasoft X rays, which preferentially ionize the K shell of carbon atoms (Hervé du Penhoat et al., Radiat. Res. 151, 649-658, 1999), was used to investigate the location of sensitive sites for cell inactivation and the physical modes of action of radiation. The enhancement of the RBE above the carbon K-shell edge either may indicate a high intrinsic efficiency of carbon K-shell ionizations (due, for example, to a specific physical or chemical effect) or may be related to the preferential localization of these ionizations on the DNA. The second interpretation would indicate a strong local (within 3 nm) action of K-shell ionizations and consequently the importance of a direct mechanism for radiation lethality (without excluding an action in conjunction with an indirect component). To distinguish between these two hypotheses, the efficiencies of core ionizations in DNA atoms (phosphorus L-shell, carbon K-shell, and oxygen K-shell ionizations) to induce damages were investigated by measuring their capacities to produce DNA double-strand breaks (DSBs). The effect of photoionizations in isolated DNA was studied using pBS plasmids in a partially hydrated state. No enhancement of the efficiency of DSB induction by carbon K-shell ionizations compared to oxygen K-shell ionizations was found, supporting the hypothesis that it is the localization of these carbon K-shell events on DNA which gives to the 340 eV photons their high killing efficiency. In agreement with this interpretation, cell inactivation and DSB induction, which do not appear to be correlated when expressed in terms of yields per unit dose in the sample, exhibit a rather good correlation when expressed in terms of efficiencies per core event in the DNA. These results suggest that core ionizations in DNA, through core-hole relaxation in conjunction with localized effects of spatially correlated secondary and Auger electrons, may be the major critical events for cell inactivation, and that the resulting DSBs (or a constant fraction of these DSBs) may be a major class of unrepairable lesions.

Electron microscopic visualization of N-acetoxy-N-2-acetylaminofluorene binding sites in ColE1 DNA by means of specific antibodies

ColE1 DNA has been allowed to react in vitro with N-acetoxy-N-2-[14C]acetylaminofluorene in the range of 0-15 N-2-[14C]acetylaminofluorene residues bound per molecule of DNA, at the C8 of guanine residues. Purified rabbit antibodies to both N-2-(guanosine-8-yl)-acetylaminofluorene and native DNA that had reacted with N-acetoxy-N-2-acetylaminofluorene were shown by electron microscopy to recognize specifically the acetylaminofluorene-modified ColE1 DNA. The antibodies bound to DNA were visualized either per se or after reaction with goat anti-rabbit immunoglobulins coupled with ferritin. There was a linear relationship between the average number of antibodies bound per DNA molecule and the number of N-2-(deoxyguanosine-8yl)-acetylaminofluorene residues per DNA molecule. The slope of this straight line was equal to 0.4. Due to the bivalence of the immunoglobulins one would expect a value of 0.5; we actually observed an important fraction of the bound antibodies crosslinking two parts of the same (or of another) DNA molecule.

Dewar valence isomers, the third type of environmentally relevant DNA photoproducts induced by solar radiation

UV-induced DNA damage is the main initiating event in solar carcinogenesis. UV radiation is known to induce pyrimidine dimers in DNA, including cyclobutane dimers and (6-4) photoproducts which have been extensively studied. In contrast, much less attention has been paid to Dewar valence isomers, the photoisomerisation product of (6-4) photoproducts. Yet, the available data show that Dewar isomers can be produced by exposure to sunlight and may lead to mutations. Dewars are thus environmentally and biologically relevant. The present review summarizes currently available information on the formation, mutagenic properties and repair of this class of UV-induced DNA damage.

8-Methoxypsoralen induced mutations are highly targeted at crosslinkable sites of photoaddition on the non-transcribed strand of a mammalian chromosomal gene

We have determined the mutational specificity of 8-methoxypsoralen photoaddition at the endogenous adenine phosphoribosyltransferase gene of Chinese hamster ovary cells hemizygous for this locus. In addition, the distribution of 8-methoxypsoralen photo-adducts was resolved in vitro at the DNA sequence level, and compared with the observed site specificity for mutation. Among 27 mutants characterized, all were single base changes at AT base pairs: 16 A:T-->T:A, six A:T-->C:G, four A:T-->G:C and one -T frameshift. All these vents were targeted to potential sites of photoaddition. The vast majority of these sites were also detectable in vitro, suggesting that 8-methoxypsoralen plus UVA-induced mutational hotspots may be damage hotspots. Furthermore 26/27 mutations occurred at crosslinkable 5'TpA sites, supporting the notion that 8-methoxypsoralen biadducts rather than monoadducts are major premutagenic lesions in mammalian cells. Since 90% of our mutation collection could have resulted from damage on the non-transcribed strand, it appears that photoadducted thymine residues on the transcribed strand of the adenine phosphoribosyltransferase gene may be preferentially repaired. We therefore suggest a model for mutagenesis, induced by psoralen biadducts, based on the preferential incision of biadducts followed by translesion synthesis past modified T bases persisting on the non-transcribed strand.

New aspects of the repair and genotoxicity of psoralen photoinduced lesions in DNA

Several approaches are described aiming at a better understanding of the genotoxicity of psoralen photoinduced lesions in DNA. Psoralens can photoinduce different types of photolesions including 3,4- and 4',5'-monoadducts and interstrand cross-links, oxidative damage (in the case of 3-carbethoxypsoralen (3-CPs)) and even pyrimidine dimers (in the case of 7-methylpyrido(3,4-c)psoralen (MePyPs)). The characterization and detection of different types of lesions has been essential for the analysis of their possible contributions to genotoxicity. For example, oxidative damage photoinduced by 3-CPs can be detected by the formamidopyrimidine glycosylase (FPG) protein. Furthermore, it is shown how the presence of MePyPs induced monoadducts may interfere with the photoreactivation of concomitantly induced pyrimidine dimers, how the ratio of monoadducts and interstrand cross-links (CL) affects the occurrence of double-strand breaks during the repair of photolesions and genotoxicity. In vitro treatment of yeast plasmids, followed by transformation, also indicates that the repair of photoadducts on exogenous DNA differs for 8-methoxy-psoralen (8-MOP) induced mono- and diadducts and for monoadducts alone. The recombinational rad52 dependent pathway is not needed for the repair of 8-MOP induced monoadducts. The results obtained suggest that the genotoxic effects of psoralens are conditioned by the nature, number, ratio and sequence distribution of the photolesions induced in DNA.

Damage distribution and mutation spectrum: the case of 8-methoxypsoralen plus UVA in mammalian cells

We determined the distribution of monoadducts and biadducts induced in the supF tRNA gene carried by the shuttle vector pZ189, after exposure to 8-methoxypsoralen (8-MOP) plus a double UVA (365 nm) irradiation. These data were compared to our previously published 8-MOP-photoinduced mutation spectrum obtained after propagation of the damaged shuttle vector in mammalian cells. One mutational hot spot in an ATAT/TATA sequence is targeted at a hot spot of biaddition. A second hot spot is not related to the presence of photoadducts either at or near the site. Moreover, it is located in a sequence which can be defined as 'mutation-prone'. Mutations occurring at GC base pairs are not targeted at sites of photoaddition, and may result from a decrease in fidelity of DNA polymerase when copying the damaged vector.

Reactivity of the antibodies to DNA modified by the carcinogen N-acetoxy-N-acetyl-2-aminofluorene

Rabbits were immunized with native DNA modified by the carcinogen N-acetoxy-N-acetyl-2-aminofluorene. The interactions between the purified antibodies to nDNA-AAF (or the Fab fragments) and several ligands have been studied. By radioimmunoassay, nDNA-AAF, dDNA-AAF, and GMP-AAF were found to bind to the antibodies with about the same affinity. GMP-AF interacts slightly less, and GMP and N-OH-AAF do not interact. The values of the association constants deduced from fluorescence measurements for the binding of the Fab fragments to nDNA-AAF, dDNA-AAF, and GMP-AAF, in 50 mM NaCl, pH 7.5, are of the same order of magnitude. The values of the association constants with nDNA-AAF and dDNA-AAF depend upon salt concentration. From this variation; it is deduced that 1-1.5 phosphate groups interact by charge--charge interactions with the Fab fragments. The absorption and circular dischroism spectra of GMP-AAF, nDNA-AAF, and dDNA-AAF bound to the Fab fragments show that the Fab fragments induce similar perturbation to the three ligands. These results lead to the conclusion that the immunodeterminant group is the dGMP-AAF residue.