skip to main content columnskip to left navigation

Appalachian Student Research Forum

Office of Research and Sponsored Programs

Division V

Abstracts Submitted:Division V - Residents & Fellows


SWITCH THERAPY WITH ATAZANAVIR: EFFECT ON LIPIDS IN FOUR PATIENTS WITH HYPERLIDEMIA ON HAART

Halim S.Aboufaycal MD, Costy Mattar MD, Jonathan P. Moorman MD, PHD. Department of Medicine, James H. Quillen College of Medicine, East Tennessee State University, Medical Service, James H. Quillen VAMC Johnson City, Tennessee 37604.

Hyperlipidemia in the setting of highly active antiretroviral treatment for human immunodeficiency virus infection is common and therapeutically challenging. While closely associated with protease inhibitors use, other agents including nucleoside reverse transcriptase inhibitors have induced lipid abnormalities. A new protease inhibitor, atazanavir, has been reported to be free of the hyperlipidemic effects of protease inhibitors. We retrospectively examined the effect of switching from traditional protease inhibitors to atazanavir on lipid profiles of four patients with hyperlipidemia in the setting of highly active antiretroviral treatment. We found fairly dramatic reductions in cholesterol and triglycerides in the majority of these patients. Given the difficulty with treating hyperlipidemia in the setting of highly active antiretroviral treatment, switch therapy may offer the most reasonable approach to this clinical problem.


Identification of a membrane bound 2-keto-3-deoxy-D-manno-octulosonic acid (KDO)-hydrolase in Helicobacter pylori

D. M. Milhorn 1, A. X. Tran 1, S. McGrath 2 , R. Cotter 2 and M. S. Trent 1; 1Depart. of Micro., James H. Quillen Coll. of Med. ETSU, Johnson City, TN, 2Johns Hopkins Univ. School of Med., Baltimore, MD.

Lipopolysaccharide (LPS) is the major surface molecule of Gram-negative bacteria and consists of three distinct structural domains: O-antigen, core, and lipid A. The lipid A domain anchors LPS to the outer membrane and is typically a disaccharide of glucosamine that is both polyacylated and phosphorylated. The core and O-antigen carbohydrate domains are linked to the lipid A moiety through the eight-carbon sugar 3-deoxy-D- manno-octulosonic acid known as Kdo. H. pylori LPS has been characterized as having a single Kdo residue attached to lipid A, predicting in vivo a monofunctional Kdo transferase (WaaA, formerly KdtA). However, using an in vitro assay system we demonstrate that H. pylori WaaA is a bifunctional enzyme transferring two Kdo sugars to the tetra-acylated lipid A precursor, lipid IV A. These findings led to the possibility of an unidentified Kdo-hydrolase acting on either mature H. pylori LPS or one of its precursors. Utilizing a number of radioactive-labeled lipid A precursors, extracts from H. pylori strains J99 and 26695 were assayed for possible modifying enzymes followed by thin layer chromatography. The results show the presence of a membrane bound enzyme that converts specific lipid A precursors to a more hydrophobic reaction product. The activity was dependent upon the presence of the Kdo 2 moiety and prior removal of the 1-phosphate from the lipid A domain. Interestingly, the LPS of H. pylori is also modified by a specific phosphatase involved in the modification of the 1-phosphate of its lipid A. Mass spectrometry analysis of the reaction product confirmed the enzymatic removal of a single Kdo residue. This is the first report of a Kdo hydrolase involved in the modification of Gram-negative bacterial LPS.


SEVERE BILATERAL UPPER EXTREMITY THROMBOPHLEBITIS RELATED TO INTRAVENOUS AMIODARONE: A RARE COMPLICATION; CASE REPORT AND REVIEW OF THE LITERATURE.

Kais Albalbissi, MD, Omar Aljitawi, MD, Bahaeddin Shabaneh, MD, and Jack Whitaker, MD, FACC, Israel Garcia, MD, Division of Cardiology - Department of Internal Medicine, James H. Quillen College of Medicine, East Tennessee State University, and James H. Quillen, VA Medical Center, Johnson City, TN

Intravenous amiodarone has been widely used since it was introduced twenty years ago for severe intractable arrhythmias. Superficial thrombophlebitis was frequently noted in the early case reports when high dose intravenous amiodarone was used. We report a 47-year-old male who developed severe bilateral upper extremity thrombophlebitis following use of intravenous amiodarone for sustained ventricular tachycardia complicating myocardial infarction. We have reviewed the literature for similar cases, and tried to reach to some conclusions related to this problem. Superficial thrombophlebitis could extend hospitalization and become a significant source of distress to our patients. Some authors recommended insertion of central line to administer intravenous amiodarone especially with expected extended use of therapy. The treating physician should be vigilant and switch from intravenous therapy to oral therapy as soon as the patients condition stabilizes and can begin oral therapy.


ACUTE MYOCARDIAL INFARCTION WITH NORMAL CORONARY ANGIOGRAPHY: AN UNUSUAL PRESENTATION OF ACUTE MYELOGENOUS LEUKEMIA.

Kais Albalbissi, MD, Omar Aljitawi, MD, Bahaeddin Shabaneh, MD, and Jack Whitaker, MD, FACC, Israel Garcia, MD, Division of Cardiology, Department of Internal Medicine, James H. Quillen College of Medicine, East Tennessee State University

Acute myeloid leukemia is a well-known hematopoeitic disorder, that usually presents with symptoms and signs of bone marrow insufficiency, such as anemia, fever, infections and bleeding. Rarely, acute myocardial infarction is the presenting diagnosis for acute myeloid leukemia. Here we report a case of a 52-year-old male patient who presented with acute myeloid leukemia and acute myocardial infarction, and found to have normal coronary angiogram. We will discuss the possible causes of acute myocardial infarction with normal coronary angiography in the setting of acute myeloid leukemia. We believe that this case is the first time in English-language literature that an acute myeloid leukemia is presented with acute myocardial infarction with normal cardiac catheterization.


Treating Amiodarone induced thyrotoxicosis with radioactive iodine: a case report and review of the literature.

Said B. Iskandar MD, Richard M. Jordan MD, Alan N. Peiris MD. Division of Endocrinology, Department of Internal Medicine, East Tennessee State University, Johnson City, TN, 37604

Amiodarone is an antiarrythmic agent widely used for the treatment of tachyarrythmias, but is also used to treat paroxysmal supraventricular tachycardia and atrial fibrillation and to maintain normal sinus rhythm after cardioversion for atrial fibrillation. A wide range of side effects can be seen, but thyroid dysfunction is probably the most recognized. In 14-18% of patient treated with amiodarone, there is overt thyroid dysfunction, either amiodarone induced hypothyroidism(1-32%) or amiodarone induced thyrotoxicosis (AIT: type I and II)(1-23%). Controversy exist regarding the utility of using radioactive iodine uptake (RAIU) in this situation. Martino et al. have suggested a normal to high RAIU in patients with preexisting thyroid gland abnormalities whereas Bogozzi et al. and Wiersinga et al. suggested that in the United States (USA), the RAIU is almost always low. This was attributed to the high iodine intake in the USA. The American association of clinical endocrinologists (AACE) recommendations are currently not to undergo RAIU in patients with AIT. Multiple studies in recent years though had shown, a decrease in iodine intake in the United States, which make us wonder whether the above statement needs to be revised.

We describe a patient with amiodarone induced thyrotoxicosis with unusual features. To our knowledge, this would be the first reported case of amiodarone induced thyrotoxicosis presenting initially as Grave's disease associated with a high radio-active iodine uptake (RAIU), with a successful outcome following treatment with radioactive iodine. Radioactive iodine was used successfully to treat amiodarone induced thyrotoxicosis in this patient. Preferential T3 secretion is a characteristic of Grave's disease. Eighty seven percent of patients with Grave's disease have a serum T3 to T4 ratio greater then 20 compared to only 15% in patients with destructive thyroiditis and 6% of euthyroid people. This ratio has not been studied in AIT but a high value would argue in favor of AIT type I. Our patients' initial T3 to T4 ratio of 121, associated to a positive TSI and diffuse enlargement of his thyroid gland on RAIU favored the diagnosis of Grave's disease. To our knowledge, this is the first reported case of Grave's disease in the setting of amiodarone use treated successfully with radioactive iodine.

Amiodarone induced thyrotoxicosis can be refractory to medical management and sometimes necessitates a total thyroidectomy. The role of radioiodine therapy has been traditionally dismissed. Whether all patients with AIT should have the radioiodine thyroid scan is still uncertain and needs further studies. We propose that in the presence of prior thyroid disease or in areas of relatively low iodine ingestion it would be prudent to obtain an I 123 thyroid scan and uptake since normal to elevated readings may offer, in selected patients, the therapeutic use of I 131.


ANTENATAL MANAGEMENT OF TRISOMY 21 PREGNANCIES: A SURVEY OF THE SOCIETY FOR MATERNAL-FETAL MEDICINE.

Nikki Yarbrough, D.O. and William Block, M.D., Department of Obstetrics and Gynecology, East Tennessee State University, College of Medicine, Johnson City, TN 37614

Down Syndrome (Trisomy 21) can now be antenatally diagnosed with either chorionic villus sampling or genetic amniocentesis. It has been well recognized that fetal aneuploidy is complicated by pregnancy loss. The fetal loss rate of Trisomy 21pregnancies is reported to be up to 43% between the time of chorionic villus sampling and term, and as high as 23% between the time of amniocentesis and term. With this in mind, the question arose: How do we manage known Trisomy 21pregnancies? Attempts were made to research this topic yet no controlled trials of antenatal surveillance strategies to reduce stillbirth rates have been reported. We therefore decided to investigate this question. To determine if any national or regional practice patterns have developed to address this issue, we conducted a survey of the current members of The Society for Maternal-Fetal Medicine. Each SMFM member was surveyed with the following questions: What are your recommendations to a mother carrying a known Trisomy 21 fetus? What is your current antenatal management of a documented Trisomy 21 pregnancy? If you do antenatal testing, at what gestational age do you begin testing? If during pregnancy there are no signs of fetal distress or IUGR, timing of delivery is what? Does your management of Trisomy 21 differ from that of other aneuploidies? A total of 1793 surveys were mailed. Currently, 615 responses have been received. Various practice patterns, recommendations and physician demographics are presented.


TRACHEOESOPHAGEAL FISTULA AND LARYNGOMALACIA- WHAT IS THE LINK?

Michelle Taylor, M.D. and Des Bharti, M.D., Department of Pediatrics, East Tennessee State University, College of Medicine, Johnson City, 37614

The most common congenital anomaly of the esophagus is esophageal atresia and affects approximately 1 in 4000 neonates. Ninety percent of these affected neonates have associated tracheoesophageal fistula, a condition in which there is a communication between the esophagus and the trachea. Patients often present in the first hours of life with increased secretions, coughing, choking, cyanosis, and respiratory distress. Feeding often worsens these symptoms, and the inability to pass a nasogastric ororogastric tube down the esophagus is usually diagnostic. There are five common forms of esophageal atresia; four having associated fistulous connections. The condition can also be associated with laryngomalacia as well as VATER syndrome. In this comparison study, we examined two patients with atypical presentations of tracheoesophageal fistula (TEF). One patient was noted to have TEF and laryngomalacia, however, his type of TEF is not normally associated with laryngomalacia. The other patient with TEF did not have laryngomalacia, even though his type is normally associated with laryngomalacia. We concluded that there are multiple factors associated with the development of laryngomalacia in children with TEF, including the position of the fistulous connection and the success of the surgical reanastamosis of the esophagus.


DISCOVERY OF A C-REACTIVE PROTEIN MUTANT THAT DOES NOT BIND TO LOW-DENSITY LIPOPROTEIN

Sanjay K. Singh, M. V. Suresh, Antonio E. Rusinol and Alok Agrawal

Department of pharmacology and Department of Biochemistry & Molecular Biology, ETSU, Johnson City, TN - 37614

Human C-reactive protein (CRP) is an acute phase protein whose production by hepatocytes is induced under inflammatory conditions. Elevated serum CRP level indicates the onset or the presence of inflammatory diseases such as atherosclerosis. Immunohistochemistry has shown deposition of CRP in the atherosclerotic lesions and it is felt that CRP may play a role in the development of such lesions. The major binding specificity of CRP is for phosphocholine (PCh)-containing ligands including oxidized low-density lipoprotein (oxLDL). The goal of the present study was to define the binding site on CRP for oxLDL. We hypothesized that the PCh-binding site of CRP might be involved in binding to oxLDL because PCh was found to inhibit CRP-oxLDL interaction. Based on the crystal structure of CRP-PCh complex, we previously employed site-directed mutagenesis to mutate the PCh-binding site and generated a CRP mutant (F66A) incapable of binding to PCh. To test the hypothesis that the PCh-binding site of CRP is also involved in binding to oxLDL, we compared the ox-LDL-binding activity of native and F66A mutant CRP. In an ELISA-based CRP-oxLDL binding assay that we developed for this purpose, native CRP bound to oxLDL in a dose-dependent manner while the F66A mutant did not bind oxLDL at all. We conclude that the PCh-binding site of CRP participates in the interaction with oxLDL and that Phe66 in CRP is critical for binding to both PCh and oxLDL. For our ongoing experiments employing ApoE knockout mice to investigate the functions of passively administered human CRP in atherosclerosis, the F66A CRP incapable of binding to PCh and oxLDL is a precious mutant.


MECHANISMS OF INTERACTION BETWEEN C-REACTIVE PROTEIN AND FIBRONECTIN

M. V. Suresh, Sanjay K. Singh and Alok Agrawal, Department of Pharmacology, ETSU, Johnson City, 37614

Most functions of C-reactive protein (CRP) are attributed to its ability to interact with phosphocholine (PCh)-containing substances; an interaction that requires prior binding of 2 calcium ions (Ca) to CRP. CRP also binds to extracellular matrix proteins such as fibronectin (Fn) through undefined mechanisms and with undefined functional consequences. CRP-Fn interaction does not require Ca and occurs in a pH-dependent manner with maximal interaction at pH 5.0. We employed site-directed mutagenesis to probe the Fn-binding site and to produce a CRP mutant incapable of binding to Fn that could help investigate the roles of CRP-Fn complexes. Since, under physiological conditions, Ca inhibits the binding of CRP to Fn, we hypothesized that the Fn-site might consist of amino acids in CRP that co-ordinate Ca. The 2 Ca in CRP are co-ordinated by Asp60, Asn61, and by residues (Glu138, Gln139, Asp140, Glu147 and Gln150) in a loop. Accordingly, seven CRP cDNAs encoding D60A; N61A; E138A; Q139A; D140A, E147A and Q150A mutants were constructed and then COS cells were transfected with the cDNAs to express the proteins. Only 2 mutants, D60A and E147A, were successfully expressed. In the Fn-binding assay at pH 7.0, both mutants failed to bind Fn supporting our hypothesis that the Ca-site contributes to the structure of the Fn-site on CRP. Surprisingly, in the Fn-assay at pH 5.0, both the CRP mutants were found to retain their ability to bind Fn. Furthermore, in contrast to the findings at pH 7.0, Ca did not inhibit the binding of either native or the 2 CRP mutants to Fn at pH 5.0. The ability of the mutants D60A and E147A CRP to bind Fn at pH 5.0 and the inability of Ca to inhibit CRP-Fn interaction at pH 5.0 indicate that under these conditions, CRP does not bind to Fn through its Ca-site. These results raise the possibility of the presence of a second Fn-site on CRP. We propose that CRP interacts with Fn by 2 mechanisms: at pH 7.0 through the Ca-site, and at pH 5.0 through an as yet unknown site.


CYSTIC FIBROSIS CARRIER SCREENING: GOOD TECHNOLOGY WITH LIMITED CLINICAL UTILITY

Jodi McCartney, M.D., East Tennessee State University

In 2002, the American College of Obstetrics andGynecology recommended that all obstetrical patients be offered a screening test to detect carriers of Cystic Fibrosis. This test can pick up a large percentage of the most common genetic mutations that cause Cystic Fibrosis. It is important to note that great progress has been made in the treatment of CF so that it is no longer the death sentence that it used to be. Further, there is no gene therapy available to treat a fetus determined to have CF in utero. Moreover, despite a reported carrier rate of 1:22 caucasians, few patients have elected to be screened. We decided to survey our patients to determine why this technology is not being utilized. Data is being collected and analyzed at this time.


CUTANEOUS VASCULITIS, A RARE PRESENTATION OF HAIRY CELL LEUKEMIA AND DRAMATIC RESPONSE TO CLADRIBINE: CASE REPORT.

Omar Aljitawi MD
, Vijay Ramu MD, Anand Karnad MD. Department of Internal Medicine, Quillen College of Medicine, East Tennessee State University, Johnson City, TN 37614.

Hairy cell leukemia (HCL) is a rare chronic B-cell lymphoproliferative disorder. First it was described under the term "leukemic reticuloendotheliosis" by Bouroncle and colleagues in 1958, and later named HCL by Schrek and Donnelly in the 1960s because of the prominent irregular cytoplasmic projections of the malignant cells. Vasculitis as a complication of lymphoproliferative diseases is relatively uncommon. Based on clinical presentation and histopathologic findings, the vasculitis observed in the patients with HCL can be classified into three distinct groups. These are leucocytoclastic vasculitis, PAN and vessel wall invasion by hairy cells. Here we report an 82 year old male with past medical history of diabetes mellitus who presented with non-healing ulcer of his left leg and pruritic rash on his extremities and abdomen. Examination revealed maculo-papular-purpuric rash which was noted on all extremities and abdomen. The left anterior leg showed erythema with some scaling/desquamation. The rest of his exam was unremarkable except for hepatosplenomegally. Laboratory examination showed leucocytosis (WBC was 13.8), which was predominantly lymphocytic (61.5%). Blood smear showed predominately "hairy" lymphocytes and smudge cells. Bone marrow was reported as slightly hypercellular marrow (approximately 60% cellularity), due to a diffuse, interstitial proliferation of lymphocytes, consistent with a chronic lymphoproliferative disorder. Flow cytometry of bone marrow aspirate was consistent with diagnosis of HCL with positive CD19, CD20, CD22, CD11C, FMC-7, KAPPA light chains. At that time, the diagnosis of drug reaction rash was challenged and the more compelling diagnosis was vasculitis related to hairy cell leukemia. This assumption was confirmed by skin biopsy which showed urticarial vasculitis. Cultures, hepatitis B serology, and HIV testing were negative. ESR was 5; urinalysis was negative for blood or protein. Ultrasound of abdomen confirmed Hepatosplenomegaly. Given the significant symptoms that patient had he was counseled about use of cladribine 2-CDA inhibitor for symptomatic vasculitis. The patient had standard 7 day continuous infusion of cladribine which he completed uneventfully. Upon follow up, his CBC normalized in 4 days. Three weeks following infusion the rash completely disappeared. Reviewing the literature, this case represents the second reported case of cutaneous vasculitis that responded to treatment with cladribie.


CONGENITAL GASTROSCHISIS IN THE NORTHEAST TENNESSEE PERINATAL REGION - AN EPIDEMIOLOGIC PERSPECTIVE

Moncenya Chatman M.D. and Elizabeth Pryor M.D., Department of OB/GYN, East Tennessee State University, College of Medicine, Johnson City, TN. 37614

Congenital Gastroschisis is a malformation in the upper anterior abdominal wall of a developing fetus. The defect is adjacent and invariably to the right of the umbilical cord insertion. While prognosis is overall good, the cause of this abnormality remains uncertain. As the incidence is seemingly increasing worldwide, investigation of associations and similarities among cases becomes paramount. The purpose of this study was to review the population incidence of congenital gastroschisis in the Northeast Tennessee Perinatal Region over the past decade. A population-based incidence study of gastroschisis from 1993 to 2003. Maternal and perinatal outcome data were collected to determine incidence and similarities in treatment and outcome trends.


CORONARY ECTASIA: A RARE MANIFESTATION OF ATHEROSCLEROSIS WITH UNSOLVED QUESTIONS.

Vipul Brahmbhatt, Baha Shabaneh, MD, Steve Smith, MD, Jack Whitaker,MD,FACC.East Tennessee State University, Johnson City, TN-37601.

Coronary artery ectasia (CAE) is defined as a luminal diameter of epicardial coronary artery>1.5 times that of a normal segment. It is also characterized by diffuse, irregular, saccular, or fusiform dilatation of the coronary artery. Although the underlying mechanisms are not fully understood, CAE is considered an atypical form of vascular remodeling in response to atherosclerosis. It is estimated that atherosclerosis presents in 50% of the cases. It has also been associated with vasculitides (Polyarteritis nodusa and Takayasus disease), trauma, and hyperlipidemia in adult population and Kawasaki disease and coronary fistula in pediatric age group. We discuss a patient with diffuse coronary artery ectasia who presented with recurrent exertional chest pain. Coronary artery ectasia was diagnosed on cardiac catheterization.

The patient is a 42-year Egyptian male with multiple risk factors for coronary atherosclerosis including gender, hypertension, diabetes, hyperlipidemia, current smoking and family history of early coronary artery disease. He presented to the hospital with exertional chest pain on and off for one year. He had no history of drug abuse. His chest pain was typically exertional in nature, retrosternal in location and without radiation. It was associated with diaphoresis, aggravated by exertion and relieved by rest. His chest pain became more frequent and more severe for few days prior to presentation. The only outpatient medication was aspirin 325mg/day. Physical examination was unremarkable except for mild obesity and mildly elevated blood pressure. Electrocardiogram showed normal sinus rhythm with rate of 76 and no ST-T changes or q wave. He was admitted and ruled out for acute myocardial infarction. EKG remained unchanged and he was chest pain free throughout the hospitalization. Considering a high pre-test probability for coronary artery disease, he was scheduled for cardiac catheterization that showed diffuse ectasia of his left circumflex artery (10 mm in diameter), left anterior discending artery (6mm in diameter). His left circumflex artery showed intracoronary swirling effect of the contrast suggestive of sluggish flow. Left main artery was normal at 4mm. No significant stenosis was seen in any coronary artery. He was discharged with recommendations for risk factor modification and Aspirin. Nitroglycerine was avoided due to the possible risk of worsening symptoms in patient with coronary artery ectasia. Although rare, Coronary artery ectasia presents a diagnostic and therapeutic challenge. It also carries an increase risk of myocardial infarction. Risk factor modification and possibly anticoagulation with coumadin could be beneficial but further investigation is warranted.


IDENTIFICATION OF POTENTIAL IRON-REGULATED GENES IN THE HUMAN GENITAL PATHOGEN, CHLAMYDIA TRACHOMATIS

Annette Rau, Susan Wyllie, Judy Whittimore and Jane E. Raulston, Departments of Microbiology and Pathology, East Tennessee State University, College of Medicine, Johnson City, TN 37614

As an obligate intracellular bacterium the genital pathogen Chlamydia trachomatis is highly responsive to the nutritional status of the eukaryotic host cell. Colonizing human endometrial epithelial cells challenge this bacterium particularly in menstruating women because of dramatic variations in iron availability. Iron is well known to play an important role in various biological processes. Therefore it is crucial for chlamydial survival to directly respond to these environmental changes raising the need for a tight regulation. In many bacteria, iron-responsive proteins are regulated at the level of transcription by a family of repressors resembling the E. coli Ferric uptake-regulator (Fur) protein. Previous studies in our laboratory discovered DcrA (Divalent cation-dependent regulator A), a distantly related homolog of E. coli Fur, representing the first repressor identified in this intracellular organism (Wyllie and Raulston, 2001). The findings suggest a Fur-like regulatory mechanism also exists in Chlamydia. Furthermore, performing a heterologous screen using E. coli Fur (Fur titration assay) we obtained approximately 5,000 recombinant plasmids having chlamydial genomic inserts that interact with E. coli Fur. The purpose of this study was to select clones from a group of 50 plasmids (1% of the library) displaying the highest level of binding affinity and to test which of these chlamydial sequences are in fact recognized and bound by chlamydial DcrA. Five out of five DNA sequences showed binding activity to DcrA when tested in electrophoretic mobility shift assays (EMSA). Four sequences are localized within the chlamydial genome less than 300 base pairs upstream of predicted translational start codons whereas the fifth sequence maps to a region less than 100 base pairs downstream of a predicted open reading frame (ORF). The ORFs are coding for genes involved in energy metabolism (CT248), shape determination (CT709) and membrane transport mechanisms (CT597). Particularly, CT248 and CT709 may be reasonable targets for iron regulation by a Fur-like mechanism in Chlamydia.


MECHANISMS OF INTERACTION BETWEEN C-REACTIVE PROTEIN AND FIBRONECTIN

M. V. Suresh, Sanjay K. Singh and Alok Agrawal, Department of Pharmacology, ETSU, Johnson City, 37614

Most functions of C-reactive protein (CRP) are attributed to its ability to interact with phosphocholine (PCh)-containing substances; an interaction that requires prior binding of 2 calcium ions (Ca) to CRP. CRP also binds to extracellular matrix proteins such as fibronectin (Fn) through undefined mechanisms and with undefined functional consequences. CRP-Fn interaction does not require Ca and occurs in a pH-dependent manner with maximal interaction at pH 5.0. We employed site-directed mutagenesis to probe the Fn-binding site and to produce a CRP mutant incapable of binding to Fn that could help investigate the roles of CRP-Fn complexes. Since, under physiological conditions, Ca inhibits the binding of CRP to Fn, we hypothesized that the Fn-site might consist of amino acids in CRP that co-ordinate Ca. The 2 Ca in CRP are co-ordinated by Asp60, Asn61, and by residues (Glu138, Gln139, Asp140, Glu147 and Gln150) in a loop. Accordingly, seven CRP cDNAs encoding D60A; N61A; E138A; Q139A; D140A, E147A and Q150A mutants were constructed and then COS cells were transfected with the cDNAs to express the proteins. Only 2 mutants, D60A and E147A, were successfully expressed. In the Fn-binding assay at pH 7.0, both mutants failed to bind Fn supporting our hypothesis that the Ca-site contributes to the structure of the Fn-site on CRP. Surprisingly, in the Fn-assay at pH 5.0, both the CRP mutants were found to retain their ability to bind Fn. Furthermore, in contrast to the findings at pH 7.0, Ca did not inhibit the binding of either native or the 2 CRP mutants to Fn at pH 5.0. The ability of the mutants D60A and E147A CRP to bind Fn at pH 5.0 and the inability of Ca to inhibit CRP-Fn interaction at pH 5.0 indicate that under these conditions, CRP does not bind to Fn through its Ca-site. These results raise the possibility of the presence of a second Fn-site on CRP. We propose that CRP interacts with Fn by 2 mechanisms: at pH 7.0 through the Ca-site, and at pH 5.0 through an as yet unknown site.


icon for left menu icon for right menu