Why do community health workers volunteer? A qualitative study in Kenya.

OBJECTIVES: Globally, there is a dire shortage of healthcare workers, but the situation is particularly bad in low- and middle-income countries. To address this, task shifting of clinical work to lower-level staff, including volunteer health workers, has been used. Whilst there are examples worldwide of such an approach working, the sustainability of programmes based on a volunteer workforce is less certain. In addition, little is known about the factors that motivate such volunteers. This study sought to ascertain these motivational drivers. STUDY DESIGN: Qualitative study using focus group discussions. METHODS: Qualitative study of volunteer community health workers (CHWs) in a rural district of Western Kenya. Twenty-three CHWs were sampled purposively, and took part in six focus group discussions. Thematic analysis was performed on the transcribed discussions. RESULTS: A variety of factors were identified as important drivers of motivation. These included financial as well as non-financial drivers, such as personal recognition, personal development and working conditions. CONCLUSIONS: There are serious unanswered questions regarding the viability of healthcare programmes founded on a workforce reliant on volunteer CHWs. This study revealed the importance of some form of reward, be it financial or otherwise, in order to retain and maintain the engagement and motivation of volunteer CHWs in these settings.

[Proposal for a new method for evaluation of physical exercise capacity in patients with chronic respiratory disease].

We developed a new method of evaluating the tolerance for physical exercise in patients with chronic respiratory disease. Using a newly developed portable pulse oxymeter, with which we could measure kinetic energy (physical activity), calculated from the vertical acceleration involved in physical movements in the patient's daily life, we considered the correlation between the characteristics of the distribution of oxygen saturation (SpO2) and the degree of physical activity. The characteristics of SpO2 distribution in normal healthy subjects are uniform at all degrees of physical activity. In patients with chronic respiratory disease who complained of dyspnea on exertion, these characteristics became more uneven as physical activity increased. By comparing the characteristics of SpO2 distributions at rest with those at certain degrees of physical activity, we could quantitatively evaluate the exercise tolerance of patients with chronic respiratory disease, while monitoring their physical activity and SpO2 in daily life, without burdening the patients with stress such as would be imposed by the treadmill test. This new method is applicable for determining the indications for home oxygen therapy. Its application in home health care could offer a useful evaluation of a patient's activities of daily living and also early discovery of aggravation of chronic respiratory failure.

A case of advanced esophageal cancer showing a long-term complete response with chemotherapy with nedaplatin alone.

We describe a case of advanced esophageal cancer treated successfully by chemotherapy with nedaplatin alone. A 60-year-old male with type 2 advanced esophageal cancer, which was located in the upper part of the esophagus and had invaded adjacent organs, was treated with nedaplatin 150 mg/body (100 mg/m2) given intravenously every 4 weeks from January 6, 1991. He achieved a partial response (PR) and was discharged in March 1991. Subsequently, he received nedaplatin 75 mg/body in an out-patient setting almost every month until August 1992. Toxicities were tolerable and included mild thrombocytopenia and nausea/vomiting. From serial evaluation in October 1993, the esophageal tumor was not observed. After 7 years since initial chemotherapy was administered, he still survives without the disease.

Fatal hepatic failure caused by miliary tuberculosis in a hemodialysis patient: case report.

A 57 year-old male dialysis patient died soon after the onset of high fever, hypoglycemia, and disturbance of consciousness. Autopsy revealed granulomatous lesions associated with caseous necrosis mainly found in the liver, despite the absence of pulmonary changes on chest radiographs performed during the patients illness. It appears that tubercle bacilli were hematogenously disseminated mainly to the liver causing miliary tuberculosis without producing typical diffuse lesions in the lungs. Since tuberculosis is a common complication in hemodialysis patients, the potential development of atypical miliary tuberculosis should always be borne in mind.

Displacement of gold marker in immunoelectron microscopy of human respiratory cilia.

Preembedding immunogold electron microscopy was performed to evaluate the position of outer arm dynein heavy chains in normal human respiratory cilia. Anti-dynein antibody (AD2), which is specific for sea urchin sperm flagellar dynein heavy chains, was used as primary antibody. Direct cross-sections of cilia were selected, and the distance between the center of a cilium and the center of a colloidal gold particle attached to the cilium (X) was measured. The distance between the center of a cilium and the farthest edge of an outer dynein arm of the cilium was measured by ordinary electron microscopy (Yo) and by immunoelectron microscopy (Yi). X was significantly longer than Yo and Yi. If it is assumed that the structure of respiratory cilia is dense and that antibodies are located at the outer side of the actual position of the heavy chains, then the average distance difference of approximately 90-120 A may represent the length of two conjugated antibodies. This length should be kept in mind when performing immunoelectron microscopy. The data suggest that AD2 recognizes the outer arm dynein heavy chains of normal human respiratory cilia.

Inhomogeneities of ventilation and the diffusing capacity to perfusion in various chronic lung diseases.

Although impairment of gas exchange caused by ventilation-perfusion (VA/Q) mismatch has been extensively analyzed, there have been no systematic studies focused on determining the distributions of diffusion properties in dose connection with those of VA/Q. We attempted to clarify the simultaneous distributions of VA/Q and diffusion capacity to perfusion (D/Q) in patients with idiopathic pulmonary fibrosis (IPF) or chronic obstructive pulmonary disease (COPD). To assess pathologic determinants causing functional abnormalities, we compared VA/Q and D/Q distributions with the findings on high-resolution computed tomography. O2, CO2, and CO together with six foreign inert gases were used as indicator gases. We transformed the measured data on indicator gases in arterial blood into a continuous distribution of Q in the VA/Q-D/Q field. In IPF, active alveolitis or acinitis played a major role in producing low D/Q regions impeding gas exchange via a diffusion limitation, whereas extensive fibrosis with minimal inflammation accounted for low D/Q as well as low VA/Q regions. In COPD, no regions with low D/Q ratios were observed, but an abnormality in the VA/Q distribution with low or high VA/Q ratios was identified. Emphysematous lesions produced high VA/Q regions, whereas peripheral airway involvement yielded low VA/Q regions. These findings suggest that hypoxemia in patients with IPF is caused by inhomogeneous distributions of D/Q in combination with those of VA/Q. Hypoxemia in patients with COPD is attributable primarily to inhomogeneities in VA/Q rather than in D/Q distributions.

Effect of hyperoxia on adhesion molecule expression in human endothelial cells and neutrophils.

To investigate the pathogenesis of pulmonary oxygen toxicity, we examined the effect of hyperoxia on adhesion molecule expression in cultured human pulmonary artery endothelial cells (HPAEC) and human umbilical vein endothelial cells (HUVEC). Endothelial cell monolayers were exposed to either hyperoxic (90% O(2)-5% CO(2)) or normoxic (21% O(2)-5% CO(2)) conditions for various periods. The level of intercellular adhesion molecule (ICAM)-1 expression had increased in hyperoxia-exposed HPAEC and HUVEC at 48 h (194 +/- 38 and 233 +/- 56%, respectively; P < 0.001) and at 72 h (200 +/- 43 and 223 +/- 52%, respectively; P < 0.001) compared with normoxic conditions. These hyperoxia-induced ICAM-1 expressions were dose dependently attenuated by a protein kinase C inhibitor (H-7). In contrast, the levels of P-selectin and E-selectin expression in HPAEC and HUVEC were unchanged. The levels of ICAM-1 mRNA and the numbers of adherent neutrophils were increased in HPAEC and HUVEC at 48 and 72 h of hyperoxia. On the other hand, hyperoxia caused neutrophil H(2)O(2) production without affecting the level of CD11/CD18 expression. These results suggest that increased ICAM-1 expression in endothelial cells plays an important role in neutrophil accumulation during hyperoxia.

[Effects of hyperoxia on the behavior of leukocytes in rat pulmonary microcirculation assessed by confocal laser scanning microscopy].

To study the dynamic interaction between blood cells and endothelial cells in the pulmonary microcirculation, we developed a method for observing leukocytes and erythrocytes in the microcirculation of perfused rat lungs by confocal laser scanning microscopy. We examined the behavior of leukocytes in the microcirculation of rat lungs exposed to hyperoxia, because oxygen toxicity is known to be associated with leukocyte infiltration and endothelial cell damage. Rats were divided into two groups: control (21% O2) and hyperoxia (90% O2 for 48 hours). Lungs were perfused with Krebs-Henseleit solution equilibrated with 21% O2 and 5% CO2 through an artificial lung. Leukocytes stained with carboxyfluorescein diacetate succinimidyl ester and erythrocytes stained with fluorescein isothiocyanate were added to the perfusate, and images of the cells were observed and recorded with a confocal laser scanning microscope and a high-speed video camera. The mean velocities of erythrocytes (Vr) in arterioles, capillaries, and venules of the control group were 1.52, 0.50 and 1.61 mm/sec, respectively. Also in the control group, the velocities of the leukocytes were divided by the mean velocities of the erythrocytes in the same arterioles, capillaries and venules (Vw/Vr) and the results were 0.96, 0.97, and 0.98, respectively. The Vw/Vr values for arterioles in the hyperoxia group were not significantly different from those in the control group, but the Vw/Vr values for capillaries and venules were 27% and 37% lower than their respective control values. Leukocyte sequestration was seen mainly in capillaries in the hyperoxic group. These results suggest that an increase in adhesion in capillaries and venules, such as that caused by adhesion molecules, might play a key role in the behavior of leukocytes in the pulmonary microcirculation during hyperoxia.

Behavior of stimulated leukocytes in the pulmonary microcirculation of perfused rat lungs.

To investigate the dynamics of activated leukocytes and the roles of CD18-ICAM-1 pathway, we examined the effects of rat IL-8 and monoclonal antibodies (mAbs) against CD18 and ICAM-1 on the behavior of leukocytes in microvessels of perfused rat lungs. Specific pathogen free male Sprague-Dawley rats were used. Perfused rat lungs were prepared so as to obtain stable physiological shear rates. We used a confocal laser scanning microscope equipped with a high speed video analysis system to visualize pulmonary microcirculation. Rat leukocytes were activated with rat IL-8. No rolling leukocytes were observed in either pulmonary arterioles or venules, and leukocytes were sequestered in capillaries. The majority of unstimulated capillary leukocytes moved smoothly. About 50% of stimulated leukocytes, however, showed a transient cessation of movement in pulmonary capillaries. Rat IL-8 decreased the relative leukocyte velocities against mean blood velocities in capillaries (45%) and venules (65%), and increased intracapillary neutrophils. Anti-CD18 and anti-ICAM-1 mAbs attenuated these changes. These results suggest that unique features exist in the interaction between activated leukocytes and pulmonary microvessels, and that CD18-ICAM-1-dependent capillary sequestration is one of the major mechanisms by which activated leukocytes accumulate in lungs.

Temporal alterations of endothelial-vasodilator functions in lung injury induced by monocrotaline.

To assess the time course of alterations in pulmonary endothelial vasodilator functions during pathological development of pulmonary hypertension (PH) and right ventricular hypertrophy (RVH), we examined changes in serotonin (5-HT) removal rates and the production of prostacyclin (PGI2) and nitric oxide (NO) in isolated rat lungs harvested at various times after single exposure to monocrotaline (MCT). We assessed the generation of vasodilator substances under conditions of both the absence and the presence of 5-HT in lungs perfused with blood-free solution. Major findings included: (i) remodeling of the pulmonary vasculature associated with RVH evident 14 days after MCT injection; (ii) the capacity for 5-HT removal was suppressed at day 1 and 7 but had been restored by day 14 after MCT exposure; (iii) basal PGI2 production in the absence of 5-HT was augmented at day 1 but had returned to control levels in lungs harvested 7 or 14 days postinjection of MCT; (iv) PGI2 production evoked by 5-HT was suppressed in MCT lungs obtained at all time points examined; (v) basal NO production was suppressed at day 1 but enhanced at day 7 and 14 in MCT lungs; (vi) NO production elicited by 5-HT stimulation in 1-day-MCT lungs was obviously suppressed while that in 7- and 14-day-MCT lungs had been restored to the control level. These findings suggest that transitional changes in endothelial functions including 5-HT removal and production of vasodilators in MCT lungs do not follow the same time course.

Clinical presentation of pulmonary tuberculosis associated with acquired immunodeficiency syndrome in metropolitan Tokyo.

The clinical features of pulmonary tuberculosis associated with acquired immunodeficiency syndrome (AIDS) in Japan were surveyed utilizing questionnaires completed by 48 institutes around the Tokyo metropolitan area. We found 11 Japanese and six foreign patients. The average number of patients per institute was 0.37. The Japanese patients had advanced human immunodeficiency virus (HIV) infection. A middle aged man, with fever and cough, nonspecific chest X-ray infiltrates, decreased lymphocyte counts, and a negative tuberculin skin test was the typical presentation of the Japanese patients. The clinical diagnosis was often difficult. The smear positive rate was high among those from whom smears were obtained, suggesting high communicability. None of the isolates indicated multidrug-resistant tuberculosis at the time of diagnosis. In conclusion, sputum smear and culture remain the keys to diagnosing tuberculosis in patients with AIDS, although the clinical picture may be atypical for pulmonary tuberculosis.

[Two cases of pulmonary disease caused by Mycobacterium chelonae subsp. abscessus].

We encountered two-cases of pulmonary disease caused by M. chelonae subsp. abscessus, [Case 1] A 72-year-old man was admitted to the hospital because of fever. He had been observed for one year after being given a diagnosis of pulmonary disease caused by Myocobacterium avium complex. Sputum examination revealed acid-fast bacilli (Gaffky 9). He recovered after administration of clarithromycin (CAM) and other drugs. [Case 2] A 61-year-old man was admitted to the hospital because of coughing and sputum production. He had been observed for 4 years after being given a diagnosis of pulmonary M. fortuitum disease. Sputum examination revealed acid-fast bacilli (Gaffky 7). His symptoms deteriorated even though he received anti-tuberculosis agents and CAM. After measurement of minimal inhibitory concentration (MIC), he was given amikacin (AMK). In both cases, the bacilli found in sputum obtained on admission were identified as M. chelonae subsp. abscessus by DNA hybridization. They were completely resistant to all anti-tuberculosis agents. However, the disk method show that they were sensitive to AMK, imipenem and CAM. The MIC value of those strains to CAM was 0.78 microgram/ml in case I and more than 100 micrograms/ml in case 2. The results obtained by MIC measurement were consistent with the clinical outcome. AMK, cefoxitin (CFX), and CAM had been used to treat M. chelouae subsp. abscessus in Europe, but the MIC value differed from strain to strain within a species. Thus the present data suggest that measurement of the MIC value of CAM would be necessary to predict its therapeutic effect.

Regulation of IFN-alpha/beta genes: evidence for a dual function of the transcription factor complex ISGF3 in the production and action of IFN-alpha/beta.

BACKGROUND: Efficient production of interferons (IFNs) in virally infected cells is an essential aspect of the host defence. The transcription factor complex ISGF3 (IFN-stimulated gene factor 3) was originally identified as a critical mediator of the IFN signal; it is formed upon IFN receptor (IFNR) stimulation and binds to ISREs (IFN-stimulated response elements) to activate IFN-inducible genes. It has recently been shown that the DNA binding component of ISGF3, p48 (ISGF3gamma) also binds to virus-inducible elements in the IFN-alpha/beta genes, suggesting a potential new role of p48 in IFN production. RESULTS: Primary cells from mice with a targeted disruption of the p48 gene show severe defects in virus-induced IFN-alpha/beta gene expression. A similar defect was also observed in cells lacking type I IFNR or Stat1, further demonstrating the role of IFN signalling in the induction of these IFN genes. ISGF3 in fact binds to the virus-inducible elements within the IFN-alpha/beta promoters. We also provide evidence showing that these elements are additionally controlled by an unidentified factor(s) which presumably triggers the primary phase of IFN gene induction. CONCLUSIONS: Our results demonstrate that the IFN signal transducing complex ISGF3 plays a crucial role in IFN production and suggest that ISGF3 may participate directly in the activation of IFN-alpha/beta promoters. This dual function of ISGF3 may insure the efficient operation of this cytokine system in the host defence.

Modulation of ICAM-1 expression by extracellular glutathione in hyperoxia-exposed human pulmonary artery endothelial cells.

To investigate the mechanisms regulating hyperoxia-induced intercellular adhesion molecule-1 (ICAM-1) expression, we studied the effects of antioxidants on ICAM-1 expression, and the relationship between ICAM-1 expression and extracellular glutathione levels in human pulmonary artery endothelial cells (HPAEC) and human umbilical vein endothelial cells (HUVEC). Cells were cultured to confluence and exposed to hyperoxia (90% O2) for 48 h with or without various antioxidants, including superoxide dismutase (SOD), catalase, N-acetylcysteine (NAC), and glutathione. The levels of ICAM-1 expression in the endothelial cells and the concentrations of reduced (GSH) and oxidized glutathione (GSSG) in the media were examined by flow cytometry and spectrophotometry, respectively. After exposure to hyperoxia, ICAM-1 expression was increased, and the supernatant total glutathione was decreased as compared with those at normoxia. SOD did not change ICAM-1 expression. The hyperoxia-induced increase in ICAM-1 expression was even greater with the addition of catalase. The ICAM-1 expression was decreased and the GSH concentration was increased with the addition of NAC. There were negative relationships between the level of ICAM-1 expression and the supernatant total glutathione concentration in catalase-treated HPAEC (R = 0.822, P < 0.0005) and HUVEC (R = 0.567, P < 0.01). Negative relationships were also demonstrated between the level of ICAM-1 expression and the total extracellular glutathione concentrations in NAC-treated HPAEC (R = 0.877, P < 0.0005) and HUVEC (R = 0.727, P < 0.0005). Exogenous GSH decreased ICAM-1 expression in both hyperoxia-exposed HPAEC and HUVEC, while exogenous GSSG did not. These results suggest that extracellular GSH plays a role in regulating hyperoxia-induced ICAM-1 expression in HPAEC and HUVEC.

[Abnormalitis in the distributions of ventilation-perfusion ratios and diffusing capacity-perfusion ratios in three types of chronic obstructive pulmonary disease].

To assess whether diffusion-limited gas exchange plays a significant role in hypoxemia in various types of chronic obstructive pulmonary disease(COPD), we analyzed the distribution of ventilation-perfusion (VA/Q) ratios and of diffusing capacity-perfusion (G/Q) ratios. We compared VA/Q and G/Q distribution in patients with three basic types of COPD: emphysematous changes, bronchiolar involvement, and airway hypersecretion, which were classified based on symptoms and on findings of high-resolution CT. The results were that 1) hypoxemia was not caused by diffusion-limited gas exchange with low G/Q regions in any type of COPD, that 2) hypoxemia was not caused by inhomogeneities in VA/Q distribution, that 3) emphysematous changes and bronchiolar involvement were associated with high and low VA/Q regions, respectively, and that 4) either hypersecretion itself or related airway abnormalities may cause low VA/Q regions to form.

[A case of AIDS with bronchial tuberculosis].

A case is 48 years-old Japanese man who had a history of frequent sexual contact with prostitutes in Thailand and the Philippines. He presented with chief complaint of chest discomfort in April 1995. His chest X-ray film showed right mediastinal lymph node swelling in other hospital and the sputum smear was strongly positive for acid fast bacilli. In May 1995, he was admitted to our hospital and serological tests for HIV were positive both by EIA and Western blot methods. The CD4 lymphocyte count was 167/microliters. He was diagnosed as a case of AIDS according to the criteria proposed by the AIDS surveillance committee of the Japanese Ministry of Health and Welfare. Although numerous tubercule bacilli were detected in sputum, the chest X-ray did not show abnormal shadow in lung fields. So the diagnosis of bronchial tuberculosis was suspected by these apparently contradictory findings and the bronchoscopy was performed. Biopsy specimen of the bronchial mucous membrane obtained by bronchoscopy confirmed the presence of acid fast bacilli by Ziehl-Neelsen's staining method, however, histological findings were atypical of tuberculosis. A month after the initiation of treatment with isoniazid, rifampicin and ethambutol and AZT, his clinical symptoms improved and the sputum smear and the culture tests for tubercule bacilli converted to negative. Complications of AIDS, (Pneumocystis carinii infection, Cytomegalo virus infection, Kaposi's sarcoma, etc) other than tuberculosis have not developed to date. In the past reports, we could not find reports of bronchial tuberculosis with AIDS. Tuberculous granuloma formation was scarce in this case, and it was suspected that bronchial tuberculosis with AIDS would show characteristic sign as same as pulmonary tuberculosis with AIDS.

Modulation of hypoxic pulmonary vasoconstriction by antioxidant enzymes in red blood cells.

To determine whether antioxidant mechanisms within red blood cells (RBCs) significantly contribute to preserving hypoxic pulmonary vasoconstriction (HPV) in both the absence and the presence of oxidative stress, we investigated HPV changes in isolated rabbit lungs perfused with solutions containing RBCs treated with various inhibitors of superoxide dismutase (SOD), anion channels, catalase (CAT), or glutathione peroxidase (GSH-Px). Perfusion was maintained at a constant flow rate of 70 ml/min, and lung temperature at 37 to 38 degrees C. Hematocrit was adjusted to 7%. In the absence of overt oxidative stress, HPV was significantly enhanced in the perfusate containing control RBCs (untreated RBCs) as compared with that in Krebs-Henseleit buffer. Inhibition of SOD, CAT, and GSH-Px within RBCs, as well as anion channels located on the RBC membrane, had little influence on HPV. Neither exogenous SOD nor CAT altered HPV. In the presence of high levels of reactive oxygen species (ROS), generated by addition of xanthine (100 microM) and xanthine oxidase (10 mU/ml) to the reservoir, HPV was considerably suppressed in the perfusate containing only buffer but was restored in the perfusate with control RBCs. Inhibition of CAT or GSH-Px in RBCs preserved the HPV, whereas inhibition of SOD or anion channels failed to preserve HPV obtained during exposure to high ROS levels. Addition of exogenous SOD, but not CAT, to the perfusate containing RBCs in which endogenous SOD had been inhibited restored HPV under high ROS conditions. In conclusion, (1) although RBCs augment HPV in the absence of ROS, this finding is not attributable to the antioxidants in RBCs. (2) RBCs restore HPV upon exposure to high ROS. This finding may well be explained by antioxidant mechanisms operating within RBCs, especially those of endogenous SOD.

Endothelial modulation of pH-dependent pressor response in isolated perfused rabbit lungs.

With the use of isolated perfused rabbit lungs (n = 152), roles of endothelium-derived relaxing factor (EDRF) in pulmonary vascular responses to hypocapnia and hypercapnia were studied. Lungs were ventilated with a gas mixture containing 1, 5, or 10% CO2 and 21% O2, adjusting the perfusate pH to 7.8, 7.4, or 7.1, respectively. Methemoglobin (MetHb), hemoglobin (Hb), methylene blue (MB), and L-argininosuccinic acid (L-ASA) were used as modulators of EDRF. To eliminate augmented shear stress, we used papaverine during hypercapnia. As a measure of EDRF, we spectrophotometrically examined nitric oxide (NO) metabolites in the perfusate. Hypocapnia and hypercapnia evoked, respectively, unsustainable vasodilatation and vasoconstriction. Hb, MB, and L-ASA, but not MetHb, produced an increase in baseline pulmonary arterial pressure (Ppa). These agents also exacerbated vasoconstriction during hypercapnia. Hypercapnia and hypocapnia caused an increase and decrease, respectively, in EDRF production. L-ASA suppressed EDRF production in hypercapnic lungs. Papaverine did not suppress EDRF production under hypercapnia. In conclusion, 1) the effects of pH on pulmonary circulation are transient, 2) the increase in Ppa caused by hypercapnia is modulated by EDRF, and 3) the pulmonary EDRF genesis is activated by hypercapnic acidosis but suppressed by hypocapnic alkalosis.

Essential and non-redundant roles of p48 (ISGF3 gamma) and IRF-1 in both type I and type II interferon responses, as revealed by gene targeting studies.

BACKGROUND: Interferons (IFNs) are a class of cytokines which confer cellular resistance against viral infections. Type I (IFN-alpha and -beta) and type II (IFN-gamma) IFNs utilize distinct receptors, the stimulation of which results in the induction of downstream target genes. These target genes usually contain within their promoter region an IFN responsive element, termed ISRE (IFN stimulated response element) which binds a heterotrimeric transcription factor, ISGF3 (IFN-stimulated gene factor 3) consisting of p48 (ISGF3 gamma), Stat1 (Signal transducers and activators of transcription-1; alpha or beta), and Stat2. The ISRE sequence overlaps with that of IRF-E which binds another IFN-inducible factor, IRF-1 (IFN regulatory factor-1). RESULTS: We generated mice lacking p48 by gene targeting. We show that p48 plays an essential role in both type I and type II IFN responses; activation of IFN-inducible genes and establishment of the antiviral state by IFN-alpha or -gamma are both severely impaired, and ISRE-binding activities induced by both IFNs are absent in the p48-negative embryonic fibroblasts (EFs). Furthermore, we generated mice deficient for both p48 and IRF-1 and found that at least one IFN-inducible gene is dependent on both factors. CONCLUSIONS: p48 and IRF-1 do not perform redundant functions in the cell, but rather complement one another in both type I and II IFN responses.