Influence of diagnostic and treatment factors in the population pharmacokinetics of gentamicin

STUDY OBJECTIVES: To determine the feasibility of macroscopic visualization of small ovarian cancer metastases in vivo by fluorescence after intravenous administration of 5-aminolevulinic acid (ALA); to assess the time after drug injection when fluorescence of small metastases is maximum; and to correlate macroscopic in vivo fluorescence with both microscopic ex vivo fluorescence and histologic findings. DESIGN: Controlled animal study (Canadian Task Force classification I). SETTING: University-based facility. SUBJECTS: Twenty-four healthy, female Fischer rats. INTERVENTION: Diffuse peritoneal metastatic cancer was induced in Fischer 344 rats by intraperitoneal injection of 1 million syngeneic ovarian cancer cells (NuTu-19). Four weeks after induction ALA100 mg/kg was injected intravenously, and diagnostic laparotomy was performed 1, 3, 6, or 9 hours thereafter. MEASUREMENTS and MAIN RESULTS: The peritoneal cavity was illuminated with the Wood's lamp (ultraviolet light). Fluorescence was determined by direct visualization and compared with a calibrated fluorescent disk. Tissues were collected, sectioned, and examined by fluorescence and conventional light microscopy. Within 1 to 3 hours after intravenous injection of ALA, in vivo fluorescence of tumor nodules (diameter 0.4-5.0 mm) was macroscopically visible. Tumor-free peritoneum did not show fluorescence and was significantly distinguishable from cancer nodules. Fluorescence from intestinal tissues was comparable with tumor nodules. Microscopic fluorescence analysis showed similar values for tumor nodules and peritoneum. Stained histologic specimens of peritoneal surface revealed a superficial layer of cancer cells responsible for fluorescence. The time course of the fluorescence curve in the intestine peaked twice, at 1 and 6 hours after ALA injection. Macroscopically fluorescing nodules were histology confirmed as malignant. CONCLUSIONS: Fluorescence detection of small cancer nodules after intravenous injection of ALA is feasible for nodules smaller than 0.5 mm on the peritoneum. One to 3 hours after drug injection is optimal for diagnosis of metastases.     

Intravenously injected radiolabelled fatty acids image brain tumour phospholipids in vivo: differential uptakes of palmitate, arachidonate and docosahexaenoate

This paper investigates the incorporation of intravenously (i.v.) administered radiolabelled fatty acids--[9,10(3)-H]palmitate (3H-PA), [1-14C]arachidonate (14C-AA) and [1-14C]docosahexaenoate (14C-DHA)--into intracerebrally implanted tumours in awake Fischer-344 rats. A suspension of Walker 256 carcinosarcoma tumour cells (1 x 10(6) cells) was implanted into the right cerebral hemisphere of 8- to 9-week-old rats. Seven days after implantation, the awake rat was infused i.v. for 5 min with 3H-PA (6.4 mCi/kg), 14C-AA (170 microCi/kg) or 14C-DHA (100 microCi/kg). Twenty minutes after the start of infusion, the rat was killed and coronal brain sections were obtained for quantitative autoradiography and histology. Each fatty acid showed well-demarcated incorporation into tumour tissue. Areas of necrosis or haemorrhage showed no or small levels of incorporation. The ratios of incorporation into the tumour to incorporation into contralateral brain regions were 2.8-5.5 for 3H-PA, 2.1-3.3 for 14C-AA and 1.5-2.2 for 14C-DHA. The mean ratios differed significantly between the fatty acids (P < 0.01). 3H-PA was not incorporated into necrotic tumours despite the presence of an open blood-tumour barrier, indicated by extravasated horseradish peroxidase. The incorporation rate constant of 3H-PA was similar for small intracerebral and large extracerebral tumours. The results show that 3H-PA, 14C-AA and 14C-DHA are incorporated more readily into tumour tissue than into brain, and that the increase is primarily due to increased utilization of fatty acids by tumour cells and not due to a high blood-tumour permeability. The relative increases in rates of incorporation for the different fatty acids may be related to lipid composition of the tumour and to the requirement of and specific role of these fatty acids in tumour cell growth and division.     

In vivo fluorescence kinetics of porphyrins following intravesical instillation of 5-aminolaevulinic acid in normal and tumour-bearing rat bladders

The fluorescence kinetics of protoporphyrin IX (PPIX) following intravesical instillation of 5-aminolaevulinic acid (5-ALA) have been studied in vivo in a rat bladder tumour model. 5-ALA dissolved in NaHCO3 was intravesically instilled for 60 min in tumour-bearing and normal bladders of Wistar rats. The fluorescence was excited with the violet lines of a Kr(+)-laser and recorded in vivo by means of a fibre coupled optical multichannel analyser. The fluorescence emission bands of PPIX at lambda = 636 nm and lambda = 708 nm were detected in normal and tumorous urothelium after only 30 min. The maximum fluorescence intensity was obtained in tumorous and normal urothelium 3-4 h after instillation. The ratio of the fluorescence intensity in tumorous to normal urothelium decreased continuously from four to about two during the time range of 6 h. PPIX fluorescence following 5-ALA instillation could also be observed in kidney and liver. Fluorescence from further porphyrin species with emission bands at lambda = 617 nm and lambda = 682 nm was detected in the bladder, indicating an efflux of hydrophilic porphyrins from the hepatic pathway.     

Intra-operative laser-induced photodynamic therapy in the treatment of experimental hepatic tumours

OBJECTIVE: To examine the effect of photodynamic therapy (PDT) on experimental liver tumours in rats. DESIGN: An experimental liver tumour model was used. Each of a group of rats had two tumours simultaneously inoculated into its liver. The tumour located in the left hepatic lobe was used for PDT, and the other one, in the median lobe, as a control. The haem precursor delta-amino laevulinic acid (ALA), at a dose of 30 mg/kg body weight, was injected 60 min before laser irradiation. Rats in group I received ALA through a femoral vein. Those in group II received ALA through the portal vein. Group III had an injection of ALA solution through the portal vein plus hepatic inflow occlusion. Three and 6 days after the treatment, the rats were killed, and the tumours were measured, and ultrastructural changes were examined using scanning electron microscopy. SETTING: Lund University Medical Laser Centre, Lund, Sweden. RESULTS: The mean tumour volume of the treated tumours increased by factors of 1.9, 1.5 and 1.7 in groups I, II and III, respectively, compared with the pretreatment baseline value. However, the mean tumour volume in the control tumours increased by factors of 9.5, 4.3 and 4.8 in the respective groups. Under the light microscope, marked necrosis of the treated tumour and the surrounding liver tissue was observed. Scanning electron microscopy revealed heavy damage to the cells and vessels in the treated tumour. CONCLUSION: PDT with ALA is an effective treatment modality for rat liver tumours.     

IL-1/IL-3 gene therapy of non-small cell lung cancer (NSCLC) in rats using 'cracked' adenoproducer cells

Cytokine gene therapy was studied in established L42 tumours in syngeneic rats. L42 is a transplantable non-immunogenic non-small cell lung cancer (NSCLC). Genes coding for human interleukin-1 alpha and for rat interleukin-3 beta were transferred by injecting producer cells of recombinant adenovirus vectors into the tumour in attempts to achieve high concentrations of the cytokines inside the tumor without systemic toxicity. Limited tumour growth delay was obtained with viable producer cells. For logistic reasons stocks of pooled frozen producer cells allowed intensive treatment of groups of tumour bearing rats. The cells were lysed by thawing before administration. Ten daily injections of such 'cracked' producer cells induced reproducible tumour responses. These were due to local release of cytokines, not to systemic effects. Growth retardation also occurred in contralateral tumours which were not injected. When rats carrying established tumours were vaccinated with lysates of tumours collected during treatment with 'cracked' producer cells, significant tumour growth retardation was obtained. We speculate that both cytokines, if produced at sufficiently high concentrations in tumours, induce inflammation which in turn initiates an immune response against tumours growing at a distant site. These findings seem to justify further exploration of IL-1 and IL-3 gene transfer for the treatment of cancers.     

Accumulation of 5-aminolevulinic acid-induced protoporphyrin IX in normal and neoplastic human endometrial epithelial cells

The aim of this study was to evaluate 5-Aminolevulinic acid (ALA)-induced fluorescence of normal and neoplastic endometrial epithelial cells for diagnosis and photodynamic treatment. Fluorescence of ALA-induced PpIX in vitro was measured by flow cytometry in two different human endometrial adenocarcinoma cell lines and in normal cells cultivated from fresh endometrial tissue of three premenopausal patients. The cells were analysed after incubation with different concentrations of ALA during 3, 6, or 24 hours. Both tumor cell lines showed a statistically significant higher fluorescence of PpIX than normal epithelial cells after incubation with 1 mg ALA per ml medium during 24 hours. The well-differentiated cancer cells produced significantly more PpIX than the poorly differentiated cancer cells. Relative PpIX intensity of the two cancer cell lines correlated with cell proliferation rate as measured by the doubling times of the cells. Higher accumulation of Pp IX in neoplastic endometrium compared to normal endometrial epithelial cells may provide targeted biopsies and selective photodynamic destruction of neoplastic micro-lesions.     

Production of urokinase-type plasminogen activator (u-PA) and plasminogen activator inhibitor-1 (PAI-1) in human brain tumours

We investigated the role of plasminogen activators (PAs) and their inhibitor (plasminogen activator inhibitor-1, PAI-1) in human brain tumours. The amounts of urokinase-type plasminogen activator (u-PA), tissue-type plasminogen activator (t-PA), and plasminogen activator inhibitor-1 (PAI-1), and the activity of u-PA and t-PA were determined by enzyme-linked immunosorbent assay (ELISA), and u-PA and PAI-1 were immunolocalized using monoclonal antibodies in human brain tumours and normal brain tissues. The tissues were surgically removed from 64 patients; normal brain tissue (5 cases), low-grade glioma (4 cases), high-grade glioma (17 cases), metastatic tumour (9 cases), meningioma (benign 12 cases, malignant 6 cases), acoustic schwannoma (11 cases). u-PA activity and u-PA and PAI-1 antigen levels were significantly elevated in malignant brain tumours (malignant meningiomas, high-grade gliomas, and metastatic tumours) and acoustic schwannomas but very low in benign meningiomas, low-grade gliomas and normal brain. There was no difference in t-PA antigen levels among normal and malignant tissues, however levels of t-PA activity were markedly decreased in metastastic tumours. All malignant brain tumour tissues showed positive immunostaining for u-PA and PAI-1, however, some tumour cells showed negative intensity while others showed strong intensity for these antibodies. This contrasts to the homogeneous staining pattern found in acoustic schwannoma. These findings indicate that malignancy in human brain tumours is associated with elevated levels of u-PA and PAI-1 and that an imbalance between these proteins in a micro-environment contributes (ascribes) to tumour cell invasion.     

Activation of JNK and p38 but not ERK MAP kinases in human skin cells by 5-aminolevulinate-photodynamic therapy

5-Aminolevulinate (ALA) photodynamic therapy (PDT) is being used clinically for the treatment of skin cancers. ALA is applied as a precursor of porphyrins serving as endogenous photosensitizers. Irradiation of HaCaT cells preincubated with 1 mM ALA for 24 h with red light of 570-750 nm at a dose of 4.5 J/cm2 leads to a 6-fold elevation of cellular c-Jun N-terminal kinase activity; phosphorylation of p38 mitogen-activated protein kinase (MAPK) is enhanced to a similar extent. In contrast, neither activation nor increased phosphorylation of the extracellular stimulus-regulated kinase MAPKs is detected. p38 is also phosphorylated by ALA-PDT in the human melanoma cell lines Bro and SkMel-23, applying doses that lead to 80-95% cell death after 24 h. Hence, the effects of ALA-PDT on MAPKs are similar to stresses like UV irradiation or exposure to hydrogen peroxide with respect to activation of JNK and p38 MAPKs. They are different, however, in that extracellular stimulus-regulated kinase activity is not raised by ALA-PDT. Of the 830 pmol porphyrins/mg protein that were present at 24 h in HaCaT cells, 99 pmol/mg were intracellular. When extracellular porphyrins had been removed by washing, p38 responses were retained. Thus, intracellular porphyrins synthesized from ALA are sufficient to elicit activation of p38 on photosensitization.     

The efficacy of lamotrigine in children and adolescents with refractory generalized epilepsy: a randomized, double-blind, crossover study

In the 9L rat brain tumour model the damage to tumour and normal brain by photodynamic therapy after intratumoural photosensitizer administration (intratumoural PDT) was studied. Twenty four rats received an intratumoural injection of 4 or 40 mm3 haematoporphyrin derivative (HpD, 5 mg ml-1), followed by interstitial irradiation with 20 Joule (J) (630 nm) 5 h later. For comparison, seven rats were treated with 20 Joule 24 h after an intravenous injection of 10 mg kg-1 HpD (intravenous PDT). With the chosen PDT parameters there was no important difference between the damaged areas produced by intratumoural PDT or intravenous PDT. No selective tumour kill was observed. Even though normal brain tissue was heavily damaged, vital tumour parts were still present. Intravenous PDT caused extensive diffuse damage to small blood vessels in tumour and surrounding normal brain. Intratumoural PDT was characterised by an infiltration of polymorphonuclear cells into damaged tissue, dilatation of larger blood vessels and gross haemorrhage. These results suggest an immediate vascular shutdown in the intravenous approach, while in the intratumoural approach the vasculature remained patent initially. Because of the severe side effects observed, the use of HpD seems not advisable for intratumoural PDT of brain tumours.     

Stimulation of tetrapyrrole synthesis in mammalian epithelial cells in culture by exposure to aminolaevulinic acid

Tetrapyrrole synthesis in CNCM-1221 cells exposed to 0.6 mM aminolaevulinic acid (ALA) was found to be approximately linear over a 6-h period of incubation. The rate was not significantly affected by cell density over a range of 0.015 to 0.15 x 10(6) cells cm(-2) (final cell density). Tetrapyrrole synthesis was not affected by GABA or glutamic acid in concentrations up to 6 mM and 2.72 mM respectively, suggesting that these amino acids, which are similar in structure to ALA, do not competitively inhibit the ALA uptake pathway in these cells. Pre-exposure to haem arginate (up to 100 microM) was inhibitory, presumably by suppression (through the inhibition of ALA synthase) of an endogenous component of the response. The ALA-stimulated response was not modified by co-exposure to AIA (up to 100 mg ml(-1)). Despite significant reduction of protein synthesis, the porphyrinogenic response of cells exposed to ALA was unaffected by cycloheximide (10 microg ml(-1)) or actinomycin D (10 microg ml(-1)) even when cells were preincubated with these agents for 3 h before ALA exposure. Fetal bovine serum (10%) inhibited tetrapyrrole synthesis by 30% but increased the rate of porphyrin export by cells by a factor of 1.5. The uptake of [14C]ALA was shown to be strongly influenced by the density of the cultures. In dense cultures (final cell density of approximately 0.15 x 10(6) cells cm(-2)), the ALA uptake rate was less than 0.8 compared with a maximum rate of 4.2 fmol per cell h(-1) at a cell density of 0.02 x 10(6) cells cm(-2). Since tetrapyrrole synthesis is less affected than ALA uptake by cell density, the resultant discrepancy in ALA incorporation occurring in dense cultures implies that endogenous ALA synthesis is induced in these cells. ALA uptake was not affected by cycloheximide or actinomycin D in serum-free conditions. However, fetal bovine serum decreased external ALA uptake by about 50%. This effect was abrogated by preincubation with cycloheximide.     

Evaluation of carborane-containing porphyrins as tumour targeting agents for boron neutron capture therapy

A number of carborane-containing porphyrins were administered to mice bearing subcutaneously transplanted mammary carcinomas. Administration was via serial intraperitoneal (i.p.) injections to assess their relative toxicities and tumour affinities. Three analogues of the natural porphyrin heme and four tetraphenylporphyrins (TPPs) were given at total doses of 78-245 micrograms g-1 body weight. The water-insoluble TPPs were less toxic to mice, and delivered greater amounts of boron to tumour than did the water-soluble TPPS and the heme analogues. One such compound, NiTCP-H, delivered more than 100 micrograms B g-1 to tumour tissue with a tumour:blood boron concentration ratio greater than 500:1 and a tumour: brain boron concentration ratio greater than 50:1, 4 days after the last of six i.p. injections given over 2 days. Another TPP analogue, NiTCP, delivered approximately 50 micrograms B g-1 to tumour with similar boron concentrations in normal tissues. Neither compound was toxic to mice at total doses of approximately 200 micrograms g-1 body weight. In contrast, the heme analogues were toxic and, with the exception of VCDP, delivered less boron to tumour than NiTCP and NiTCP-H. The two porphyrins with the greatest potential for application to boron neutron capture therapy (BNCT), NiTCP and NiTCP-H, yielded higher tumour:blood and tumour:brain boron concentration ratios in mice than could be achieved with p-boronophenylalanine (BPA) and sodium mercaptoundecahydrododecaborate (BSH), the compounds which are currently being used in clinical trials of BNCT in the treatment of glioblastoma. The boron delivered by each of the porphyrins tested remained in tumour tissue longer than did boron delivered by either BPA or BSH. The copper and nickel chelates of these porphyrins behave identically in vivo. The former offer the potential for imaging by 67Cu-mediated single photon emission computed tomography (SPECT) to aid BNCT treatment planning.     

Change of karyoskeleton during mammalian spermatogenesis: expression pattern of nuclear lamin C2 and its regulation

The photophysical and photochemical properties of porphyrins were profoundly changed upon addition of rhodamine 123. The Soret band of the porphyrins shifted to higher wavelengths, the fluorescence yield of the porphyrins decreased with unaltered decay rates, and their triplet state was quenched. These observations indicate a strong interaction between porphyrins and rhodamine 123 and formation of 1:1 nonfluorescent complexes, of which the binding constants were determined. Illumination of a porphyrin in the presence of rhodamine 123 resulted in the formation of a porphyrin radical cation, which could be detected with ESR spectroscopy. Quenching of the triplet state of the porphyrins by rhodamine 123 resulted in a decreased singlet oxygen yield and a decrease of the photooxidation of histidine, methionine, tyrosine, and tryptophan. However, the oxidation of thiol compounds was increased and the stoichiometry of the reaction between cysteine and oxygen changed from 2 to 3.8 mol cysteine/ mol oxygen. These results show that the presence of rhodamine 123 converted the for porphyrins prevalent energy transfer (type II) reaction to an electron transfer (type I) reaction.     

Enhancement of 5-aminolaevulinic acid-induced photodynamic therapy in normal rat colon using hydroxypyridinone iron-chelating agents

Currently, the clinical use of 5-aminolaevulinic acid (ALA)-induced protoporphyrin IX (PPIX) for photodynamic therapy (PDT) is limited by the maximum tolerated oral ALA dose (60 mg kg(-1)). This study investigates whether hydroxypyridinone iron-chelating agents can be used to enhance the tissue levels of PPIX, without increasing the administered dose of ALA. Quantitative charge-coupled device (CCD) fluorescence microscopy was employed to study PPIX fluorescence pharmacokinetics in the colon of normal Wistar rats. The iron chelator, CP94, when administered with ALA was found to produce double the PPIX fluorescence in the colonic mucosa, compared with the same dose of ALA given alone and to be more effective than the other iron chelator studied, CP20. Microspectrofluorimetric studies demonstrated that PPIX was the predominant porphyrin species present. PDT studies conducted on the colonic mucosa showed that the simultaneous administration of 100 mg kg(-1) CP94 i.v. and 50 mg kg(-1) ALA i.v. produced an area of necrosis three times larger than similar parameters without the iron-chelating agent with the same light dose. It is possible, therefore, to increase the amount of necrosis produced by ALA-induced PDT substantially, without increasing the administered dose of ALA, through the simultaneous administration of the iron-chelating agent, CP94.     

Urinary porphyrins as biological indicators of oxidative stress in the kidney. Interaction of mercury and cephaloridine

Reduced porphyrins (hexahydroporphyrins, porphyrinogens) are readily oxidized in vitro by free radicals which are known to mediate oxidative stress in tissue cells. To determine if increased urinary porphyrin concentrations may reflect oxidative stress to the kidney in vivo, we measured the urinary porphyrin content of rats treated with mercury as methyl mercury hydroxide (MMH) or cephaloridine, both nephrotoxic, oxidative stress-inducing agents. Rats exposed to MMH at 5 ppm in the drinking water for 4 weeks showed a 4-fold increase in 24-hr total urinary porphyrin content and a 1.3-fold increase in urinary malondialdehyde (MDA), an established measure of oxidative stress in vivo. Treatment with cephaloridine alone (10-500 mg/kg, i.p.) produced a dose-related increase in urinary MDA and total porphyrin levels up to 1.6 and 7 times control values, respectively. Injection of MMH-treated rats with cephaloridine (500 mg/kg) caused a synergistic (20-fold) increase in urinary porphyrin levels, but an additive (1.9-fold) increase in the MDA concentration. Studies in vitro demonstrated that cephaloridine stimulated the iron-catalyzed H2O2-dependent oxidation of porphyrinogens to porphyrins in the absence of either microsomes or mitochondria. Additionally, porphyrinogens were oxidized to porphyrins in an iron-dependent microsomal lipid peroxidation system. Moreover, porphyrinogens served as an effective antioxidant (EC50 approximately 1-2 microM) to lipid peroxidation. These results demonstrate that MMH and cephaloridine synergistically, as well as individually, promote increased oxidation of reduced porphyrins in the kidney and that this action may be mechanistically linked to oxidative stress elicited by these chemicals. Increased urinary porphyrin levels may, therefore, represent a sensitive indicator of oxidative stress in the kidney in vivo.     

Porphyrin fluorescence in plasma of various types of porphyria

Fluorescence spectra of plasma porphyrin were measured in 6 patients with the acute intermittent porphyria (AIP), 30 patients with variegate porphyria (PV), 2 patients with hereditary coproporphyria, 2 patients with porphyria cutanea tarda, and in 6 patients with erythropoietic protoporphyria (EPP). It was found that the excitation and emission wavelengths at which maximum fluorescence is seen may help to diagnose and differentiate PV and EPP. In patients with AIP spectrum characteristic for porphyria of such a type was noted in all patients during the attack of disease and in only 33% of patients in remission. Fluorescence spectrum was normal in asymptomatic family members. In variegate porphyria spectrum with a characteristic maximum of fluorescence was noted in all patients during an attack and remission, and 62% of the asymptomatic family members.     

Fluorescence of dental calculus from cats, dogs, and humans and of bacteria cultured from dental calculus

Recently we reported that feline and canine dental calculus fluoresced pink to red under long wavelength ultraviolet light due to the presence of porphyrin. Here we report the observation of such fluorescence in 30 of 30 cats, 30 of 30 dogs, and 8 of 13 supragingival samples and 5 of 5 subgingival samples of humans. The fluorescence spectra of the calculus dissolved in 9 M HCl show that it is due to three distinct metal-free porphyrins. Similar fluorescence is obtained from bacterial cultures grown from calculus deposits of cats and dogs and bacteria grown on blood agar containing hemin and vitamin K1. The results of the bacterial culture study suggest that the metal-free porphyrin is produced by bacteria in the mouth. The clinical observation of fluorescence can be used for diagnosis.     

The incorporation of various porphyrins into blood cells measured via flow cytometry, absorption and emission spectroscopy

The incorporation of the five following porphyrins: meso-tetra(4-phenyl)porphyrin (TPP); meso-tetra(4-sulfonato-phenyl)porphyrin (TPPS4); meso-tetra(4-naphthyl)porphyrin (TNP); tri-sulfo-tetra-phenyl porphyrin (TPPS3) and tetra-sulfonato-naphthyl porphyrin (TNPS4) into human blood cells was investigated using flow cytometry, and absorption and emission spectroscopy. The percentage of stained cells, measured in a fluorescence cytometer, provided information on the efficiency of incorporation of fluorescent dye molecules into different types of cells. The yield of the incorporation of a dye was dependent on the type of dye and the solvent used for cell incubation. The degree of dye aggregation and ionization varied with the incubation medium, but dye molecules incorporated into cells seemed to be restricted to those in the monomeric state, exhibiting similar fluorescence yield. Of the three sulfonated porphyrins investigated only TPPS4 was efficiently incorporated into leukocytes. In the incubation solvent, this dye was in monomeric and neutral form. TPPS3 which was also in monomeric form, practically was not incorporated into cells. TPP and TNP dissolved in 5% aqueous dimethyl sulfoxide were present mostly in aggregated forms but they penetrated the cells with a high efficiency.     

Delayed cytokine expression in rat brain following experimental contusion

Proinflammatory cytokines mediate brain injury in experimental studies. This study was undertaken to analyze the production of proinflammatory cytokines in experimental contusion. A brain contusion causing delayed edema was mimicked experimentally in rats using a weight-drop model. Intracerebral expression of the cytokines interleukin (IL)-1 beta, tumor necrosis factor-alpha (TNF alpha), IL-6, and interferon-gamma (IFN gamma) was studied by in situ hybridization and immunohistochemistry. The animals were killed at 6 hours or 1, 2, 4, 6, 8, or 16 days postinjury. In the injured area, no messenger (m)RNA expression was seen during the first 2 days after the trauma. On Days 4 to 6 posttrauma, however, strong IL-1 beta, TNF alpha, and IL-6 mRNA expression was detected in mononuclear cells surrounding the contusion. Expression of IFN gamma was not detected. Immunohistochemical double labeling confirmed the in situ hybridization results and demonstrated that mononuclear phagocytes and astrocytes produced IL-1 beta and that mainly astrocytes produced TNF alpha. The findings showed, somewhat unexpectedly, a late peak of intracerebral cytokine production in the injured area and in the contralateral corpus callosum, allowing for both local and global effects on the brain. An unexpected difference in the cellular sources of TNF alpha and IL-1 beta was detected. The cytokine pattern differs from that seen in other central nervous system inflammatory diseases and trauma models, suggesting that the intracerebral immune response is not a uniform event. The dominance of late cytokine production indicates that many cytokine effects are late events in an experimental contusion: Different pathogenic mechanisms may thus be operative at different times after brain injury.