On the Web at
w / COMIRBProtocol / PROTOCOL #:(Use Protocol Manager on the COMIRB Website)
COLORADO MULTIPLE INSTITUTIONAL REVIEW BOARD
CAMPUS BOX F-490 TELEPHONE: 303-724-1055 Fax: 303-724-0990
Project Title:CoPARC: Colorado Pulmonary Alcohol Research Consortium
Principal Investigator: Ellen L. Burnham, MD, MS
I.Hypotheses and Specific Aims
Our resource will be focused on establishingpulmonary effects of alcohol use disorders (AUDs), including alcohol abuse and dependence that contribute to an increased susceptibility for pneumonia. Based on prior investigations, we hypothesize thatalcohol-related alterations on pulmonary oxidative stress, the cytokine milieu, and endogenous proteins lead to an increased susceptibility to pneumonia through their adverse effects on alveolar macrophage (AM) and bronchial airway epithelial cell (BAEC) function, and their influence on the respiratory tract microbiome. These abnormalities may be further augmented in the setting of smoking.
Specific Aims: Determine the mechanisms whereby AUDs increase the predisposition to pneumonia via effects on
1. Alveolar macrophage (AM) efferocytosis, apoptosis, and maturation, and their relationship to:
1a. Zinc deficiency
1b. Pulmonary oxidative stress
2. Bronchial airway epithelial cell (BAEC) function, including:
2a. Expression of toll-like receptor-2 (TLR-2) and ciliary motility
2b. Response to protein adducts formed in the setting of AUDs and smoking
3. Respiratory tract microbial ecology, and its relationship to alterations in antimicrobial protein composition and function within the alveolar space
II. Background and Significance:
The effect of AUDs on the incidence and severity of pneumonia poses a substantial burden to our health care system. Pneumonia (along with influenza) is responsible for approximately 1.3 million hospitalizations and costs our economy $40.2 billion dollars annually(1). It ranks as the eighth leading cause of death, and is the most common cause of death from infectious diseases in the US(2). An estimated 50% of patients with pneumonia have an AUD(3), and these patients have more severe symptomatology, longer and more expensive hospitalizations, and higher mortality(4);(5);(6), despite their relatively younger age (<60 years old(7;8)). Among surgical patients with AUDs, an increased risk of post-operative pneumonia has similarly been described(9) in association with prolonged ventilator dependence, a longer ICU stay, and mortality(10;11). As a result, there are more AUD-associated deaths from pneumonia when compared to the number of deaths due to alcohol-related liver disease(12), traffic accidents(13), or alcohol-related pancreatitis(14).Successful completion of the proposed aims will lead to identification of therapies to decrease the incidence of alcohol-associated pneumonia.
Co-morbid factors, including smoking, can further alter susceptibility for pneumonia in the setting of AUDs.For example, investigations in animal models have revealed that ciliary dysfunction is compounded in the presence of both alcohol and cigarette smoke compared to either exposure alone(15), suggesting that a smoke/AUD interaction may adversely affect airway innate immune function (16).Prior clinical investigations have not accounted for the influence of these common co-morbid factors on pneumonia in the AUD setting. Therefore, investigations proposed will specifically do so by (1) ensuring sample sizes are large enough to explore differences between AUD/non-AUD subjects, with and without a specific confounder, (2) utilizing study designs (e.g. pair-matching) and statistical analyses (e.g. stratification by smoking) to factor in confounder effects on outcome variables.
Four other institutions that are leaders in alcohol-related lung research have expressed overwhelming interest in establishing a consortium with UCD to improve translational discoveries related to alcohol-associated pneumonias.The majority of investigations from sites outside of UCD have been conducted in animal models, although the population and infrastructure at all but one of these sites (University of Nebraska) can support the research focus of this consortium. Combining four sites’ clinical research capabilities will greatly enhance the feasibility of the proposed translational research projects by ensuring sizable numbers of human subjects, with and without AUDs. Denver is home to a large population of racially diverse AUD subjects with little other co-morbidity. Emory, the only other site presently conducting clinical alcohol-related lung research, has a largely African-American population, as does LSU, where clinical lung investigations in subjects with HIV/healthy controls are on-going. Emory and LSU have agreed to further supplement our UCD resource with samples from their sites. With a large cohort of diverse subjects, we can explore effects of AUDs on the predisposition for pneumonia while accounting for important confounders, including smoking.Our resource will create productive interactions in a forum that encourages a logical investigative pathway extending from animal models to the bedside. It will also facilitate the recruitment of new investigators interested in alcohol-associated pulmonary research.
The research paradigm for alcohol-related pneumonia has historically been weighted heavily toward animal model and cell-line based research. This paradigm has resulted in few phase III trials, and so far has failed to provide clinically relevant interventions for humans(17). The establishment of this consortium seeks to shift the standard paradigm and transform this research model by establishing a logical progression of investigations with human subjects in order to fuel the conduct and improve success of phase III clinical trials.This consortium’s proposed experiments focus on animal-to-human investigations,confirming differences in lung-specific outcome measures in individuals with AUDs compared to healthy individuals. The goals of planned experiments are (1) to explore novel hypotheses relevant to the development of pneumonia, (2) to confirm significant alcohol-related animal model abnormalities, and (3) to extend small-scale observations made previously in in vitro human investigations into a larger cohort to better account for confounders, to determine the most important mechanisms in AUDs that predispose to pneumonia. Importantly, this research will identify potential therapies todecrease the incidence of pneumonia in individuals with AUDs and other critical illnesses, such as sepsis. Robust and persistent AUD-associated alterations will be targeted for expanded investigations, while differences initially observed in animal models but not confirmed in humans will not require further interest or investigations.
The proposed research capitalizes on the expertise of a diverse group of investigators and will promote cross-site synergy and discovery in the field of alcohol-related pneumonias. By design, this R24 will facilitate frequent interactions between our sites that will encourage collaboration. For example, LSU investigators have prior experience in assessing the respiratory tract microbiome, and will establish its microbial diversity in the setting of AUDs, while UCD is focused on determining antimicrobial protein composition in the alveolar space. Collectively, these two groups’ observations could identify new antimicrobial approaches in patients with AUDs. Alternatively, alterations in zinc homeostasis assessed by investigators at Emory may ultimately prove to modulate AM maturity (as they hypothesize), and can also be explored in the context of other AM activities, including efferocytosis, one focus of UCD investigators.
III. Preliminary Studies/Progress Report:
We have successfully enrolled otherwise healthy subjects with alcohol use disorders and matched controls into protocols involving the CTRC at UCH. Since 2007, we have been recruiting subjects from Denver CARES for IRB-approved protocols examining the effects of alcohol on the lung. One of these protocols involved a 2 night observational trial including one bronchoscopy (n=50 subjects with AUDs, n=15 controls), while the other involved an 8 day clinical trial with two bronchoscopies (n=30 subjects) on the CTRC. All planned bronchoscopies were completed successfully on the unit, using conscious sedation, without adverse events. Subjects with AUDs were managed for alcohol withdrawal symptoms with the CIWA protocol; no episodes of complicated alcohol withdrawal requiring escalation of care occurred. No subjects failed to complete assigned protocols.
SA 1 [EU Project 1]: Alternative (M2) activation of AMs antagonizes M1 activation, leading to AMs that are unable to phagocytize particles but can abnormally generate reactive oxygen species, arginase, TGF-, and fibronectin (18;19). This can contribute to fibroblast proliferation with collagen production in the lung(19). Our preliminary data with NR8383 cells (a macrophage cell line), and AMs from a rat AUD model suggest that chronic ethanol treatment leads to increased expression of AM M2 activation markers. In our preliminary studies, expression of M2 activation markers was examined on human AMs via fluorescent microscopy. AUDs were associated with decreased AM expression of CD32, a marker of terminal differentiation, while expression of arginase-1, galectin-3, IL-13,mannose receptor, and TGF-β1 were increased.
SA 1a [EU Project 2]: AUDs are associated with zinc deficiency which has the potential to affect normal epithelial and immune cell functions, and lead to an increased susceptibility for pneumonia. Our group has demonstrated in a rat model that chronic alcohol ingestion is associated with altered expression of key zinc transporters and storage proteins in the lung, and decreased zinc levels in the alveolar compartment(20). In AMs from these animals, in vitro treatments with zinc increased intracellular zinc levels, promoted these cells’ terminal differentiation, and improved their bacterial phagocytic capacity. If zinc improves AM function and maturation, zinc therapy could be a logical choice to restore normal AM function in those with AUDs.
SA 1b [EU/UCD Project]: AUDs have been associated with excessive pulmonary oxidation through effects on the thiol glutathione(21);(22). However, the impact of AUDs on cysteine/cystine thiol pairs is not known. Oxidation of cysteine/cystine pairs promotes signal transduction and alters cellular responses(23;24). Therefore, it is a key factor influencing proper innate immune function. Establishing the thiol species most prominently affected by AUDs, and developing methods to measure intrapulmonary oxidation non-invasively may expedite the identification of individuals who would most benefit from anti-oxidant therapy to restore proper immune function. We have examined redox potential in non-invasively collected exhaled breath condensate (EBC) from those with AUDs (n=27) compared to controls (n=8)and determined that oxidative stress indices reflect what is found in epithelial lining fluid (ELF) within BAL(25).
SA 2a [UNMC Project 1]:TLR-2 receptors on airway epithelium recognize peptidoglycan, the cell wall component found in Gram (+) respiratory pathogens such as S. pneumoniae(26). This initiates a signaling cascade that results in inflammatory cytokine secretion. Our UNMC group reported that alcohol causes a biphasic modulation of TLR-2expression on airway epithelium, where brief alcohol exposure induces TLR-2 upregulation, while prolonged alcohol exposure causes downregulation(26). These changes result in dysfunctional inflammation. In our preliminary data, mice fed alcohol for 6 weeks (according to the Cook model(27)) express very little TLR-2 on the airway epithelium; moreover, we have observed similar changes in human airway epithelial cells linesin vitro. However, the effects of AUDs on airway epithelial cells in vivo are not known.
SA 2b [UNMC Project 2]: Investigators at UNMC have developed a well-characterized mouse model of co-exposure to alcohol and cigarette smoke(15) that we have utilized to examine reactive aldehyde protein modification, specifically malondialdehde-acetaldehyde (MAA) protein adducts to nascent lung proteins. These adducted lung proteins activate Protein Kinase C epsilon (PKCε) through their effect on scavenger receptors (SR) of lung epithelial cells (28). PKCε, in turn, affects airway epithelial cell production of pro-inflammatory cytokines in the lung(29). Human ELF from individuals with AUDs who smoke has not been specifically examined to determine MAA protein adduct quantity.
SA 3 [LSU Project]: In the setting of immunodeficiency or mucosal injury to the lung, normal interactions between the host and microorganisms are deranged, leading to abnormal microbial colonization of the lower airways. Published data from our LSU group in animal models suggests thatalcohol consumption worsens pneumonia caused by both bacterial and fungal organisms (reviewed in(30)). However, the spectrum of colonization in the respiratory tract in patients with AUDs and its contribution to disease is not defined. We contend that alcohol-induced perturbations in mucosal immunity allow specific species of the normal lung microbiota to proliferate, competitively excluding other species, unbalancing and restricting community diversity, and leading to further host immune dysfunction. This establishes a vicious cycle in which the altered microbiota leads to further dysregulation of mucosal immunity.
SA 3a [UCD Project 3]: ELFcontains a variety of antimicrobial proteins(31) that have gained additional recognition recently as immune regulators(32). Effects of AUDs on antimicrobial proteins in the respiratory tract have remained relatively unexplored, although alcohol metabolites have been demonstrated to adversely affect lysozyme, a high-abundance antibacterial protein(33).We recently demonstrated that ELF from subjects with AUDs (n=15, with n=6 controls)has significantly decreased lysozyme activity (p<0.01) and lactoferrin concentration (p<0.02). Moreover, using linear regression modeling across a range of ELF protein concentrations, killing of type 2 S. pneumoniaewas inferior by ELF from AUD subjects in an in vitro killing assay compared to ELF from controls.
IV. ResearchMethods
A. Outcome Measure(s): Measures are dependent on the specific assay as delineated in the study design and research measures, below. They include:
1. M2 activation markers on alveolar macrophages
2. alveolar macrophage zinc levels (intracellular)
3. thiol homeostasis in epithelial lining fluid and exhaled breath condensate
4. TLR2 expresssion on airway epithelial cells
5. MAA protein adducts in epithelial lining fluid
6. microbiome within lung
7. antimicrobial proteins in epithelial lining fluid
- Description of Population to be Enrolled:
Table 1. AUD Subject Inclusion/Exclusion Criteria
Inclusion criteria: Subjects will be eligible to participate in the study if they meet all of the following criteria:
1. Alcohol Use Disorders Identification Test (AUDIT) score of 8 for men and ≥5 for women.
2. Last alcohol-containing beverage consumed within the seven days prior to enrollment.
Exclusion criteria: Subjects will be ineligible to participate in the study if they meet any of the following criteria:
1. Prior medical history of liver disease: cirrhosis, total bilirubin ≥ 2.0 mg/dL, or albumin < 3.0 g/dL
2. Prior medical history of symptomatic gastrointestinal bleeding
3. Prior medical history of myocardial infarction or congestive heart failure
4. Prior medical history of end-stage renal disease or serum creatinine ≥ 3 mg/dL
5. Prior medical history of asthma
6. Concurrent illicit drug use defined as a positive urine toxicology screen for opiates and cocaine (to be performed on the CTRC prior to bronchoscopy).
7. Prior history of diabetes mellitus
8. Prior history of HIV, not controlled on medications
9. Peripheral white blood cell count of less than 3000
10.Abnormal chest radiograph or spirometry (forced vital capacity or forced expiratory volume in 1 second <60% predicted)
11. Failure of the subject or the subject’s substance abuse counselor to provide assent
12. Nutritional risk index of less than 95
13. Actively withdrawing from alcohol (defined as a CIWA score of >10)
14. Age < 18 or > 60
15. Pregnancy
Table 2. Control Subject Inclusion/Exclusion Criteria
Inclusion criteria: Subjects will be eligible to participate in the study if they meet all of the following criteria:
1. Alcohol Use Disorders Identification Test (AUDIT) score of <2 for men and <1 for women.
Exclusion criteria: Subjects will be ineligible to participate in the study if they meet any of the following criteria:
1. Prior medical history of liver disease: cirrhosis, total bilirubin ≥ 2.0 mg/dL, or albumin < 3.0 g/dL
2. Prior medical history of symptomatic gastrointestinal bleeding
3. Prior medical history of myocardial infarction or congestive heart failure
4. Prior medical history of end-stage renal disease or serum creatinine ≥ 3 mg/dL
5. Prior medical history of asthma
6. Concurrent illicit drug use defined as a positive urine toxicology screen for opiates and cocaine (to be performed on the CTRC prior to bronchoscopy).
7. Prior history of diabetes mellitus
8. Prior history of HIV positivity not controlled on medications
9. Peripheral white blood cell count of less than 3000
10. Abnormal chest radiograph or spirometry (forced vital capacity or forced expiratory volume in 1 second <60% predicted)
11. Nutritional risk index of less than 95
12. Age < 18 or > 60
13. Pregnancy
Subjects with AUDs. We have the capability of safely enrolling approximately 72 subjects per year and following these subjects to the completion of theirprotocol(s). These patients will be recruited from Denver CARES, the detoxification center associated with Denver Health. We have recruited research subjects from this facility successfully since 2008. We propose to enroll up to 360 subjects with AUDs for Years 1-5 of the proposal. Their age ranges are anticipated to be between 35 and 60, and their overall health status should be good, aside from their history of alcohol abuse. Inclusion and exclusion criteria for subjects with AUDs are listed on table 1. At the UCD site, subjects with AUDs will be recruited from Denver Health and Hospital Administration’s Denver Comprehensive Addictions Rehabilitation and Evaluation Services (Denver CARES), a local detoxification center and our collaborating site, after their discharge from this facility.At the Emory site, subjects with AUDs will be recruited from the outpatient detoxification clinic at the Atlanta VA Medical Center, and from the Grady Memorial Hospital Emergency Department population. IRB approval to recruit patients at the Emory site is in place currently.
Healthy control subjects. Approximately 250 healthy controls will also be enrolled over this same time frame (50 per year for years 1-5).
These subjects will either be recruited from the Denver VA and Atlanta VA Medical Centers’ smoking cessation clinics (smoking controls), and the pulmonary/general medicine clinics at LSU HSC. Additionally, subjects may be enrolled from the UCD, EU, and LSU medical center campuses. Their age ranges, smoking history, and gender will be reflective of the characteristics of our subjects with AUDs. Since it is possible for smoking to affect many of the lung-specific outcome variables we intend to explore, a detailed history of current and former smoking practices will be obtained from control subjects in order to match them as closely as possible to the smokers contained within the AUD subject cohort.Inclusion and exclusion criteria are listed in table 2. Of note, a positive urine toxicology screen will be performed on the CTRC for all AUD subjects and controls; evidence of recent cocaine and heroin use will automatically exclude individuals from participation.