Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An exploration of the lung microbiome and DNA methylation in patients infected with human immunodeficiency… Xu, Yan 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_september_xu_yan.pdf [ 3.78MB ]
Metadata
JSON: 24-1.0307415.json
JSON-LD: 24-1.0307415-ld.json
RDF/XML (Pretty): 24-1.0307415-rdf.xml
RDF/JSON: 24-1.0307415-rdf.json
Turtle: 24-1.0307415-turtle.txt
N-Triples: 24-1.0307415-rdf-ntriples.txt
Original Record: 24-1.0307415-source.json
Full Text
24-1.0307415-fulltext.txt
Citation
24-1.0307415.ris

Full Text

	 AN EXPLORATION OF THE LUNG MICROBIOME AND DNA METHYLATION  IN PATIENTS INFECTED WITH HUMAN IMMUNODEFICIENCY VIRUS  by Yan Xu B.Sc., The University of British Columbia, 2014  A THESIS SUBMITTED IN PARTIAL FULLFILLMENT OF  THE REQUIREMENT FOR THE DEGREE OF  MASTER OF SCIENCE in The Faculty of Graduate and Postdoctoral Studies (Experimental Medicine)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2016  © Yan Xu, 2016 	ii	Abstract	Background: With the advent of antiretroviral therapy, patients infected with human immunodeficiency virus (HIV) can achieve near normal life expectancies. However, the risk for chronic illnesses such as chronic obstructive pulmonary disease (COPD) remains higher in HIV-infected patients despite improved survival. In disease states such as COPD, the microenvironment of the lungs can change dramatically, which creates permissive niches that allow for selective growth and reproduction of microbes. These changes may influence or be influenced by epigenetic alterations, specifically the methylation and demethylation of sites along the genetic code. We hypothesize that the microbiome and methylation profiles in the lower airway of HIV-infected patients are different compared to those of uninfected patients. These changes may prove relevant to the progression of chronic lung complications. Experimental Approach: This thesis examined small airway epithelial cells collected from patients infected with HIV and from uninfected subjects. Data on bacterial microbiome were obtained by using touchdown polymerase chain reaction (PCR) on epithelial cell DNA followed by MiSeq sequencing of specific variable regions on the bacterial 16S rRNA gene. For methylation profiling, DNA samples from the same airway epithelial cells were bisulfite converted and sequenced with the Illumina Infinium 450K HumanMethylation platform. Microbial diversity calculations, and linear regression models of methylation data were analyzed on Vegan and Limma packages in R, respectively. Results: The bacterial microbiome from small airway epithelial cells within the lungs of HIV-infected patients was distinct from that in uninfected patients. These changes included decreased diversity and decreased richness. Moreover, certain operational taxonomic units (OTUs) were able to distinguish HIV from non-HIV samples. Furthermore, a significant difference in global methylation patterns was identified.  Conclusion: The microbiome and methylation profiles from small airway epithelial cells in HIV-infected individuals are significantly different from those in uninfected individuals, which may in part explain the heightened susceptibility of HIV-infected patients to COPD.  	iii	Preface	 This research was approved by the UBC-Providence Health Care Research Ethics Board (certificates number H14-03267). I performed all the microbiome experiments and data analysis as well as wrote the first draft. I am responsible for methylation sample preparation. Statistician Nick Fishbane assisted with the methylation data analysis. 	iv	Table	of	Contents	Abstract	........................................................................................................................................	ii	Preface	.........................................................................................................................................	iii	Table	of	Contents	..........................................................................................................................	iv	List	of	Tables	.................................................................................................................................	vi	List	of	Figures	...............................................................................................................................	vii	List	of	Abbreviations	....................................................................................................................	viii	Acknowledgements	........................................................................................................................	x	Dedication	.....................................................................................................................................	xi	CHAPTER	I	INTRODUCTION	............................................................................................................	1	1.1	 Human	Immunodeficiency	Virus	and	Chronic	Obstructive	Pulmonary	Disease	.....................	1	1.1.1	 Statistical	overview	of	HIV	infection	..............................................................................	1	1.1.2	 Three	decades	of	HIV/AIDS	research	.............................................................................	1	1.1.3	 Pulmonary	complications	in	the	HIV-infected	population	............................................	2	1.1.4	 The	mechanism	of	HIV-associated	COPD	......................................................................	4	1.2	 The	Bacterial	Microbiome	.....................................................................................................	5	1.2.1	 The	human	inner	ecosystem	.........................................................................................	5	1.2.2	 What	do	we	know	about	the	microbiota	in	human	disease?	........................................	6	1.2.3	 The	lung	bacterial	microbiome	......................................................................................	7	1.3	 DNA	Methylation	Profiling	....................................................................................................	8	1.3.1	 Epigenetic	crosstalk	between	host	and	the	microbes	...................................................	8	1.3.2	 DNA	methylation	in	HIV	infection	.................................................................................	9	CHAPTER	II	The	Bacterial	Lung	Microbiome	in	HIV	Infection	........................................................	11	2.1	 Introduction	.........................................................................................................................	11	2.2	 Materials	and	Methods	.......................................................................................................	14	2.2.1	 Study	population	and	sample	collection	.....................................................................	14	2.2.2	 Isolation	of	nucleic	acid	...............................................................................................	15	2.2.3	 Droplet	Digital	PCR	......................................................................................................	15	2.2.4	 MiSeq	sequencing	pipeline	..........................................................................................	16	2.2.5	 Microbiome	analysis	and	statistics	..............................................................................	17	2.3	 Results	.................................................................................................................................	18	2.3.1	 16S	rRNA	bacterial	load	analysis	.................................................................................	18	2.3.2	 Phyla	distribution	........................................................................................................	20	2.3.3	 Bacterial	diversity	analysis	...........................................................................................	22	2.3.4	 Microbial	compositional	analysis	................................................................................	25	2.4	 Discussion	............................................................................................................................	27	CHAPTER	III	Methylation	in	the	Lung	of	HIV	Infected	Population	.................................................	31		v	3.1	 Introduction	.........................................................................................................................	31	3.1	 Materials	and	Methods	.......................................................................................................	32	3.1.1	 DNA	methylation	profiling	...........................................................................................	32	3.1.2	 Data	processing	and	quality	control	............................................................................	32	3.1.3	 Statistical	analysis	........................................................................................................	34	3.1.4	 Interpreting	differences	in	DNA	methylation	..............................................................	34	3.2	 Results	.................................................................................................................................	35	3.3	 Discussion	............................................................................................................................	40	CHAPTER	IV	CONCLUSION	...........................................................................................................	43	Bibliography	................................................................................................................................	45	Appendices	.................................................................................................................................	52	Appendix	A:	An	overview	of	clinical	traits	of	HIV	infected	and	uninfected	subjects	.............................	52	Appendix	B:	DNA	extraction	protocol	for	bronchial	brushing	................................................................	55	Appendix	C:	DDPCR	primer	sequences	..................................................................................................	56	Appendix	D:	Touchdown	PCR	primer	sequences	for	16S	rRNA	gene	MiSeq	sequencing	.......................	57	Appendix	E:	Mothur	batch	code	and	final	reads	summary	....................................................................	59	Appendix	F:	Post-hoc	Tests	....................................................................................................................	62	Appendix	G:	Lists	of	GO	Pathways	from	DMRs	Enrichment	...................................................................	64				vi	1 List	of	Tables	Table	1.	Summary	of	demographics	in	HIV-infected	and	control	cohorts	.....................................	15	Table	2.Demographics	of	patients	infected	with	HIV	...................................................................	52	Table	3.	Demographics	of	patients	in	the	control	cohort	..............................................................	53	Table	4.	Forward	and	reverse	primer	sequences	for	16S	rRNA	gene	used	for	ddPCR	....................	56	Table	5.	Forward	primer	sequences	for	16S	rRNA	gene	used	for	MiSeq	sequencing	.....................	57	Table	6.	Reverse	primer	sequences	for	16S	rRNA	gene	used	for	MiSeq	sequencing	......................	58	Table	7.	Dunn’s	multiple	comparison	test	of	phylum	Bacteriodetes	.............................................	62	Table	8.	Dunn’s	multiple	comparison	test	of	phylum	Proteobacteria	...........................................	62	Table	9.	Dunn’s	multiple	comparison	test	of	phylum	Firmicutes	..................................................	62	Table	10.	Dunn’s	multiple	comparison	test	of	Shannon	Diversity	.................................................	63	Table	11.	Dunn’s	multiple	comparison	test	of	richness	................................................................	63	Table	12.	Dunn’s	multiple	comparison	Test	of	evenness	..............................................................	63	Table	13.	List	of	GO	pathways	in	biological	process	that	associated	with	HIV	infection	................	64	Table	14.	List	of	GO	pathways	in	cellular	component	that	associated	with	HIV	infection	.............	66	Table	15.	List	of	GO	pathways	in	molecular	function	that	associated	with	HIV	infection	..............	68		 	vii	List	of	Figures	Figure	1.	Vicious	cycle	in	patients	infected	with	HIV.	...................................................................	12	Figure	2.	Pipeline	to	generate	a	library	pool	for	16S	rRNA	gene	MiSeq	sequencing	......................	17	Figure	3.	16S	rRNA	bacterial	load	across	different	disease	status.	................................................	19	Figure	4.	16S	bacterial	load	between	HIV	infected	and	uninfected	subjects	.................................	20	Figure	5.	Phyla	distribution	between	HIV	positive	and	negative	population.	................................	21	Figure	6.	Phylum	distribution	across	all	groups	............................................................................	22	Figure	7.	Microbial	diversity	between	HIV	infected	and	uninfected	population.	..........................	23	Figure	8.	Microbial	diversity	across	all	groups.	.............................................................................	23	Figure	9.	Rarefaction	curve	between	HIV	positive	and	HIV	negative	subjects..	.............................	24	Figure	10.	Non-metric	multidimensional	scaling	analysis	.............................................................	25	Figure	11.	Heatmap	of	Boruta	picked	discriminative	OTUs	...........................................................	26	Figure	12.	Workflow	for	analyzing	and	interpreting	DNA	methylation	data	.................................	33	Figure	13.	β-values	for	top	6	CpG	sites	from	limma	result	top	grouped	by	HIV	status	...................	35	Figure	14.	Principle	component	analysis	plot	with	HIV	status..	....................................................	36	Figure	15.	Principle	component	analysis	plot	with	plating	information.	.......................................	37	Figure	16.	Venn	diagram	of	Limma	and	DMRcate	result..	.............................................................	38	Figure	17.	Volcano	plot	in	DNA	methylation	between	HIV+	and	HIV-	population.	........................	39	  	viii	List	of	Abbreviations	16S rRNA: 16S ribosomal DNA AIDS: acquired immunodeficiency syndrome BAL: bronchoalveolar lavage  CCR5: C-C chemokine receptor 5 CpG: cytosine-phosphate-guanine COPD: Chronic Obstructive Pulmonary Disease COPD-: without diagnosis of COPD COPD+: diagnosed with COPD CT scan: computed tomography scan ddPCR: droplet digital PCR DMRs: differentially methylated regions FEV1: forced expiratory volume in one second  FEC: forced vital capacity GO: gene ontology GOLD: Global Initiative for Chronic Obstructive Lung Disease GPR5: G-protein-coupled receptor 5 HAART: highly active antiretroviral therapy HIV: human immunodeficiency virus HIV-: patients not infected with HIV HIV+: patients infected with HIV IBD: inflammatory bowel diseases LPS: lipopolysaccharide n63: dataset with 63 samples n53: dataset with 53 samples 	ix	NMDS: non-metric multidimensional scaling NRTIs: nucleoside reverse transcriptase inhibitors OTUs: Operational Taxonomic Units  PANTHER: Protein ANalysis THrough Evolutionary Relationships PCA: principal components analysis  PCP: pneumocystis pneumonia  PCR: polymerase chain reaction perMANOVA: permutational multivariate analysis of variance PFT: pulmonary function test  SickKids: The Hospital for Sick Children SNPs: single nucleotide polymorphisms SOP: standard operating procedure (SOP) TCGA: The Centre for Applied Genomics TLR4: Toll-like receptor 4  VAC: Veteran Aging Cohort  WebGestalt: WEB-based Gene Set AnaLysis Toolkit 	 	x	Acknowledgements	 I would like to thank my research supervisor Dr. Don Sin for giving me the opportunity to work under his guidance. For two years, Dr. Sin has shown tremendous trust in me, which has led to the completion of this thesis. His advice has helped me to focus not only on my study but also my life goals. I especially thank Dr. Paul Man for bring me under his wing in the study of HIV-associated COPD. He has given me great encouragement along the way. I would like to give my sincere gratitude to Dr. Janice Leung, for always being there to guide and support me. I could not have asked for a better supervisor committee and their supervision has been the most fortunate blessing for me.  I would like offer my enduring gratitude to Dr. Marc Sze for teaching me everything I have learned about microbiome and data analysis. His great passion in microbiome research leaded every step in my research and the thesis could not be completed without his guidance. This dissertation would not have been possible without the help of my dearest Sin lab members, particularly all the technicians, Sheena Tam, David Ngan and Yeni Oh. They have provided enormous technical and moral support for the past two years.  Last but not least, I also want to thank my parents for supporting me coming this far to pursue my dreams. I am beyond blessed to be your daughter and I love you with all my heart.  	xi	Dedication	   I	DEDICATE	THIS	THESIS	TO:	 To my parents my dearest friends and the one above all    		1	1 CHAPTER	I	INTRODUCTION	1.1 Human	Immunodeficiency	Virus	and	Chronic	Obstructive	Pulmonary	Disease	1.1.1 Statistical	overview	of	HIV	infection	      In 2014, an estimated 36.9 million people were living with HIV, the majority of whom live in low- and middle income countries1. Geographically, approximately 70% of the global HIV-positive population (25.8 million) lives in sub-Saharan Africa, 5 million people live in Asia and the Pacific, and 2.4 million people live in Western and Central Europe or North American2. The United States is estimated to spend $31.7 billion dollars in 2016 for combined domestic and global HIV research, prevention and treatment. 1.1.2 Three	decades	of	HIV/AIDS	research	      The history of HIV research begins with the first reported cases of acquired immunodeficiency syndrome (AIDS) in the early 1980s3. The epidemic was first observed in homosexual men, followed by intravenous drug users, recipients of blood transfusions, and finally, the general population4. The disease soon became a devastating epidemic that spread across the world, which led to over three decades of intense study.        In the 1980s, three major milestones were achieved in HIV research. In 1984, Dr. Robert Gallo and Dr. Jay A. Levy reported the isolation of a retrovirus that they believed caused AIDS, which was later named human immunodeficiency virus (HIV)5,6. One year later, CD4+ was identified as the main HIV receptor7,8. This knowledge became the basis for monitoring CD4+ cell counts in clinical practice that still occurs today. Soon after this, the 	2	HIV genome nucleotide sequence was determined in 19859,10. Genomic information allowed scientists to identify diversity, origin and evolution of HIV, and it has also provided instrumental tools for developing viral load and resistance tests4.        A watershed moment occurred in the 1990s with the introduction of highly active antiretroviral therapy (HAART). For the first time, HAART effectively reduced viral loads and limited the development of drug resistance11. Death from AIDS has been reduced dramatically since the implementation of HAART in 199512. In the meantime, scientists continue to study the molecular biology of the virus. By the end of the twentieth century, scientists had mapped the life cycle of HIV and identified the target of multiple antiviral drugs4.        With the success of antiretroviral therapy, a 20-year-old individual who is infected with HIV and on HAART in the U.S. or Canada is now expected to live into their 70s, a life expectancy that approaches that of a 20-year-old person in the general population13. However, HAART is not curative and does not fully restore health14.  Entering the 21st century, more studies reported that patients infected with HIV have significant risks for a number of diseases typically associated with older age, including osteoporosis, cardiovascular disease, pulmonary disorders, and non-AIDS malignancies15,16. As the mechanism of this is largely unknown, researchers are now investigating how HIV infection increases the risk for seemingly unrelated non-AIDS events.   1.1.3 Pulmonary	complications	in	the	HIV-infected	population	Pulmonary disease has long been a major contributor to morbidity and mortality in patients infected with HIV. Before the advent of HAART, pneumocystis pneumonia (PCP), 	3	bacterial pneumonia, and acute bronchitis were among the most frequent complications of HIV17. Since the introduction and widespread use of HAART, the incidence of pulmonary opportunistic infection has decreased18.  Studies have found that non-infectious comorbidities in the lung, such as chronic obstructive pulmonary disease (COPD), pulmonary arterial hypertension, and lung cancer, are common in HIV-infected individuals compared to the uninfected population19. One study compared the Veteran Aging Cohort (VAC) with HIV-negative controls and found a 47% increased risk for COPD in veterans with HIV after adjusting for smoking status and age20. Another study examined patients in the San Francisco HIV/AIDS registry from 1996 to 2011, and identified increases in COPD in patients infected with HIV despite the decreasing overall mortality and AIDS-related rates21. Also, a case-control study in Africa have found COPD was 2.85 times more common in HIV-infected individuals and the mean age was 42.6 compared to 67 years in normal COPD patients22. Researchers have also studied clinic-based cohorts of individuals who underwent pulmonary function tests (PFTs) and computed tomography (CT) scans. From PFTs, they found Global Initiative for Chronic Obstructive Lung Disease (GOLD) -defined COPD in 7% -23% of patients infected with HIV, two-fold to five-fold that of uninfected individuals19,23–26. Furthermore, based on CT scans, 50% of adult patients with HIV radiographic manisfestations of COPD (either emphysema and/or bronchiolitis), compared with only 25% in the control population27. Although the mechanism of increased risk of non-infectious complications in the lungs is unclear, recent studies have shown that the manifestation may be due to a combination of causes that eventually lead to early onset of aging-associated complications in the lungs.  	4	1.1.4 The	mechanism	of	HIV-associated	COPD	      There are multiple hypotheses of how non-infectious pulmonary complications such as COPD developed in patients infected with HIV. High-risk behaviors, such as cigarette smoking and injection drug use, are proposed causes28–30. Approximately 75% of patients infected with HIV have smoked at least 100 cigarettes in their lifetime, and half of them are current smokers28. However, it is not the only explanation because pulmonary complications are seen in HIV-infected nonsmokers as well31. Moreover, epidemiologic studies have demonstrated that the risk for COPD remains elevated in HIV even after statistical adjustments for smoking histories.        Pulmonary infections and microbial colonization in patients infected with HIV may promote COPD progression. In HIV-infected patients, defective immune systems may incompletely clear microbial pathogens, increasing the likelihood of developing colonization. Colonization results in the recruitment of white blood cells as well as higher levels of inflammatory cytokines, chemokines, and proteases, which could eventually lead to tissue damage, airway thickening, and COPD31. For example, a strong association has been found between Pneumocystis colonization and COPD serverity32. Additionally, most pathogens that have been linked with HIV-associated COPD are bacteria and fungi. Smokers are more vulnerable to bacterial lung infections. Common bacterial causes of lung infection include Streptococcus pneumoniae, Haemophilus influenza, Pseudomonas aeruginosa, and other streptococci.        Altered oxidant/antioxidant balance and aberrant inflammatory responses are other possible causes for the increased risk of COPD in patients infected with HIV. Studies have 	5	found that patients infected with HIV have decreased antioxidant levels, and smoking may worsen this imbalance33. Additionally, the use of nucleoside reverse transcriptase inhibitors (NRTIs)  in HAART can also induce mitochondrial toxicity, which may result in altered oxidant/antioxidant levels34,35. The imbalance between oxidants and antioxidants leads to systemic oxidative stress, which is known to be the major etiologic factor for COPD development36. As many inflammatory mediators are sensitive to oxidative stress, the interplay between oxidative stress and inflammation in patients infected with HIV might play an important role in COPD pathogenesis. In addition, HIV itself can independently stimulate pulmonary inflammation, particularly via CD8+ lymphocytes37. CD8+ lymphocytes then secrete IFN-γ, which induces emphysema in a mouse model38.  1.2 The	Bacterial	Microbiome	1.2.1 The	human	inner	ecosystem	      Our body is home to many microbial communities. The microbes that live inside or on us outnumber our human somatic cells by a factor of ten. They reside in many body systems, including in the digestive tract, on the skin, and in the respiratory system. Interconnected microbial communities from different parts of our body vary from those living in exceptionally dry habitats on the surface of the skin to those on highly nutritious and moist surfaces like the conjunctiva of the eye and from niches of low population density, such as the lungs, to habitats with high population densities of 1012 cells/gram in the upper intestine39. When disease arises, it can change our microbial ecosystem in both the gut and lungs, despite the great differences in population density and biodiversity. Analyses of the 	6	complex species communities offer an opportunity to examine how the microbial ecosystems contribute to human health and disease. 1.2.2 What	do	we	know	about	the	microbiota	in	human	disease?		      Since the late 1800s, the cultivation and isolation of bacteria have long been the gold standard for studying microbes. However, the complex and dynamic nature of our microbiota is not fully understood because microbiological cultivation techniques are limited, as more than 70% of bacteria species within the human body cannot currently be cultivated in vitro40. Recent advances in DNA sequencing have allowed us to characterize microbial communities at a broader range and scale than ever before.        With these new molecular tools, scientists have found that human microbial community composition often changes with disease. For example, inflammatory bowel diseases (IBD) such as Crohn’s disease and ulcerative colitis, as well as type 1 and type 2 diabetes are associated with alterations in intestinal microbes41–43. On the skin, acne vulgaris and atopic dermatitis are associated with changes in the microbiota44. Gut microbiota have also been shown to influence cancer susceptibility. Although the mechanism is unclear, one study demonstrated that mice on a high-fat diet have an increased risk of developing intestinal carcinogenesis independently of obesity45.        Microbiota in the gut may have remote effects on other organ function. It has recently become evident that microbial translocation can influence the development of certain diseases. A study of the gut-brain microbiota axis has revealed how variations in the composition of gut microbiota affect the brain and behavior through communication with the central nervous system46. Microbial translocation is also found in HIV infections, whereby 	7	damage to the intestinal epithelial permits increases in microbial translocation. Higher levels of 16S ribosomal DNA (rDNA) are observed in plasma samples of patients infected with HIV47. A landmark study demonstrated that individuals infected with HIV acutely have increased microbial translocation and systemic lipopolysaccharide (LPS) levels48. Microbial product translocation elicits potent proinflammatory responses and is associated with local and systemic inflammation.  1.2.3 The	lung	bacterial	microbiome	      Historically, the lungs have been considered sterile. However, recently, diverse bacterial communities were found in the lower respiratory tract. Since the first report of the lung microbiome from Hilty et al in 2010, lung microbiome studies have rapidly increased in numbers, as over 30 publications have reported evidence of microbial communities in the lung49,50. Studying the relationship between the lung microbiome and respiratory diseases may reveal new therapeutic targets to restore microbial balance and lung health.        The microbiome in healthy lungs has been analyzed using bronchoalveolar lavage (BAL) samples in numerous studies. On the phylum level, the most common phyla normally observed are Bacteroides, Firmicutes and Proteobacteria51. The phyla characterized in BAL samples are similar to those found in the upper airway, but they differ in relative abundance51–53. When analyzed on the genus level, the common genera consistently found among healthy control populations are Prevotella, Veillonella, Streptococcus, and Pseudomonas51.         The microbiome in COPD is different compared to that in healthy controls. Hilty et al. were among the first to show an increase in Proteobacteria and a decrease in Bacteriodetes 	8	in the COPD samples49. Another study performed by Erb-Downward et al. compared the microbiota using BAL, and found that subjects with COPD who were also smokers had markedly decreased microbial diversity and overlapping community composition compared to non-smokers54. Conversely, Sze et al. found no significant difference in microbial diversity in COPD lung tissue compared to that of control lung tissue. However, a distinct microbial community was found when comparing smokers and non-smokers among COPD subjects. Firmicutes increased in COPD samples, and the shift was driven by increases in Lactobacilli55.        Overall, all studies that have compared the lung microbiome in diseased lungs with those of healthy controls suggest that there are unique bacterial communities in diseased lungs. The community composition shifts away from Bacteroidetes towards Proteobacteria in patients with COPD50. The microbial community composition changes with disease, which may affect COPD progression.   1.3 DNA	Methylation	Profiling 1.3.1 Epigenetic	crosstalk	between	host	and	the	microbes	      Effective communication is undoubtedly necessary not only in cultural relations, but also in biology. The symbiotic relationship established between two distinct living entities, humans and microbial communities, requires a shared “language” that breaks this boundary. Such communication is facilitated by epigenetic regulation. Epigenetics is the field that examines changes in gene expression that occur without changing underlying DNA sequences. In other words, it is a change in phenotype without a change in genotype56. 	9	Epigenetic modification allows different packaging and interpretation of the genome based on the influence of external or environmental factors. The microbiome can influence human epigenetic profiles directly via exposure to live microorganisms and indirectly through the secretion of metabolites and antigens such as LPS. The metabolites or antigens can then alter the activities of enzymes involved in epigenetic modification.        Crosstalk between DNA methylation and microbial colonization has been studied in animal models. One study investigated the microbial effect on the host epigenome and found that host gene methylation is influenced by LPS through Toll-like receptor 4 (TLR4)57. Another study found that microbial exposure alters gene expression in specific tissues58, with the absence of specific pathogens resulting in gene hypermethylation in colon and lung tissues. Environmental exposure later in life might trigger irritable bowel syndrome and asthma via altered methylation profiles. Overall, the microbiota has emerged as a critical factor that influences epigenetic profiles in all body sites that are colonized by microbes, and yet the mechanism remains unclear.  1.3.2 DNA	methylation	in	HIV	infection	      Not all genes are active all of the time. DNA methylation is the first and the most characterized epigenetic modification that controls gene expression. During the modification process, a methyl group is attached to a cytosine (C) residue. In adult somatic cells, methylation predominantly occurs at cytosine-phosphate-guanine (CpG) sites, and approximately 60%-90% of all CpGs are methylated59. Methylation is necessary for differentiation of different cell types because the DNA sequence is essentially the same in 	10	every cell of the human body; it is the different methylation patterns that characterize the specific tissue type.       Changes in DNA methylation at certain sites appear to be a hallmark of the aging process and have been linked to many diseases, such as HIV infection. Studies have shown that CpGs can be hypermethylated or hypomethylated in HIV-infected population compared with those uninfected subjects60–64. A predictive tool has been developed to estimate the age of human tissue and cell types using methylation, which is commonly referred to as the epigenetic clock65. The same group studied DNA methylation patterns in the brain and blood of patients infected with HIV infected and concluded that the epigenetic clock can be used as a biomarker for accelerated aging during HIV infection66. Another recent study confirmed the epigenetic aging model in patients infected with HIV by measuring whole-blood DNA methylomes of 137 HIV-positive individuals along with 44 matched controls67. They found that HIV infection lead to global changes in methylation patterns and an average aging advancement of 4.9 years. In addition, a preliminary study from our group also showed the same trend in methylation patterns from lung tissue of patients infected with HIV.    	11	2 CHAPTER	II	The	Bacterial	Lung	Microbiome	in	HIV	Infection	 2.1 Introduction       Information on the human lung microbiota has resulted in new research questions, particularly for chronic inflammatory diseases such as COPD. Although several studies have demonstrated that patients with COPD possess unique microbiome communities compared to control populations, the mechanism by which the shift in the lung microbiome influences COPD pathogenesis remains unknown49,54,55,68,69. Additionally, HIV being an independent risk factor for COPD adds another level of complexity to understanding the lung microbiome20.        To explain, the Vicious Cycle hypothesis has been proposed in the literature70–73, a theory that alterations in innate lung defense induced by environmental factors allow pathogenic bacteria to persist and proliferate. These bacteria induce inflammation by signaling though various pathogen recognition receptors on airway epithelial and immune cells. In order to clear the infection, the lung is constantly in a defense state with continuous production of inflammatory cytokines, which eventually damage the lung defense mechanisms promoting further proliferation of the pathogens. The concept applies to HIV infection as well [Figure 1]. As discussed previously, HIV infection induces chronic inflammation and oxidative stress. Destruction in lung function might change the microenvironment of the lungs, which leads to microbial imbalance or dysbiosis. In fact, a previous study has shown that declines in microbiome diversity were associated with increased alveolar destruction and greater CD4 cell infiltration74. The process, marked by increasing level of LPS and inflammatory cytokines, further promote chronic inflammation and lung destruction. Understanding the 	12	mechanism of the vicious cycle will provide insights to the role of lung microbiome in viral infections and human diseases. Figure 1. Vicious Cycle in patients infected with HIV.        Four studies have investigated the lung microbiome in HIV infection. Lozupone et al. identified unexpected colonization of Tropheryma whipplei in the BAL of HIV-positive subjects, and this colonization was reduced by effective antiretroviral therapy75. Iwai et al. compared microbiome compositions between Ugandan patients infected with HIV who also had pneumonia with similar patients in the U.S. and found that the Ugandan patients had microbiomes significantly richer and compositionally distinct from those in the US cohort76. The differences may be attributed to differences in age, clinical status, diet, ethnicity, and Chronic	 inflammationAltered	oxidant/antioxidant	balancePulmonary	damage	and	dysregulated	repairAltered	microbiome	diversity	and	compositionIncreased	level	of	LPSIncreased	level	of	inflammatory	cytokinesHIV	Infection	13	environmental exposure. Another study by Twigg et al. also demonstrated altered diversity and composition in the lung microbiome between patients with advanced HIV infection and control patients uninfected by HIV77. However, a multicenter comparison of lung and oral microbiomes of patients with and without HIV infection found no significant differences between the two groups78. The result was unexpected and might be due to the diverse background in clinical status, genetics, and environmental factors across different cohorts. All four studies were performed with BAL samples, which mainly consist of epithelial lining fluid.  This study is the first lung microbiome study that uses bronchial brushing to sample small airway epithelium cells in patients infected with HIV. The bronchial brushing method accesses cells and microbes that adhere to the luminal surface of the small airways, which might be a better representation of the colonized microbial community at the disease site. The same brushing sample is used to analyze both the bacterial microbiome and host methylation profiles. We hypothesize that the bacterial microbiome community is different in patients with HIV infected compared to uninfected individuals, and we think specific species can be identified to discriminate between the two groups. By investigating the lung microbiome in HIV infection, we will gain a better understanding on the role of microbiome in COPD pathogenesis.  In this study, we examined total bacterial 16S loads, microbial diversity and compositional profiles to evaluate overall distribution of the lung microbiome in HIV-infected and uninfected individuals. Additionally, the relative abundance of phyla was also studied in order to observe significant phylum shift during disease progression. Finally, the 	14	discriminative OTUs that drove the differences were analyzed and interpreted for biological meanings. 2.2 	Materials	and	Methods	 2.2.1 Study	population	and	sample	collection       Approval for the study was given by the UBC Providence Health Care Ethics Committee, and informed consent for the collection of cytological brushings for research purposes was obtained from each patient in writing.        Twenty-eight HIV-infected adults were recruited from patients undergoing bronchoscopy for pulmonary nodules or masses, bronchiectasis and pneumonia at St Paul’s Hospital in Vancouver, B.C., Canada. Forty-eight HIV negative adults were recruited from patients undergoing lung cancer screening at the British Columbia Cancer Agency in Vancouver. All subjects performed spirometry and had a thoracic CT scan within the last three months. Subjects with forced expiratory volume in one second (FEV1)/ forced vital capacity (FVC) ratio less than 70% were defined to have COPD79. Table 1 is a summary of demographics in this study. The detailed clinical traits of each patient can be found in Appendix A.  Cytologic brushings were collected by passing a flexible bronchoscope with minimal suction through the subject’s oropharynx. Patients were under conscious sedation during the procedure and topical 1% lidocaine anesthetic was applied throughout the vocal cords, trachea, and main bronchi. Bronchial epithelial cells were obtained from the upper lobe airways, away from areas of acute infection, masses or nodules as determined by CT scan. The brush was preserved in Cytolyt (Cytyc, Marlborough, MA) and stored at -80°C for DNA preservation. Clean brushes rinsed with distilled water were used as baseline controls. 	15	Table 1. Summary of demographics in HIV-infected and control cohorts   HIV+ (n=28)	 HIV- (n=48)	Sex	   Male 0.86	(24)	 0.52	(25)	Female 0.14	(4)	 0.48	(23)	Age 58±12	 63±7.7	FEV1/FVC	 70.06	 68.39	Smoking	Status	   Current	 0.43	(12)	 0.44	(21)	Past 0.5	(14)	 0.5	(24)	Never 0.07	(2)	 0.06	(3)	Viral	Load	Detectible	 0.32	 -	 2.2.2 Isolation	of	nucleic	acid							DNA from bronchial brushing samples was extracted using the QIAGEN DNeasy Blood & Tissue Kit (Catalog #69504). Samples were eluted with 50 ul of distilled water and the concentration was measured by nanodrop. A detailed DNA extraction protocol is outlined in Appendix B. All samples were normalized to 12 ng/ul for downstream experiments.  2.2.3 Droplet	Digital	PCR       To quantify total bacteria load in each sample, primers specifying the 293bp amplicon of the 16S rRNA gene (Table S3 in Appendix C) were designed using the protocol outlined by Sze et al80. The experiment was performed using a SYBR green based assay with the Bio-Rad QX200TM system. For each 20 ul PCR reaction, 10 ul of 2X ddPCR Supermix for Probes (Bio-Rad, Catalog #186-3010), 0.1 uM 63F primer, 0.1 uM 355R primer, 2 ul of DNA template and 7.6 ul of distilled water were used. The PCR cycling conditions were 1 cycle at 95°C for 5 minutes, 40 cycles at 95°C for 30 seconds and 60°C for 1 minute, 1 cycle at 4°C for 5 minutes and 1 cycle at 90°C for 5 minutes. All cycles were run at a ramp speed 	16	of 2°C per second. The PCR reaction was performed with a Bio-Rad T100 thermal cycler and quantified using Bio-Rad Quantisoft software. A threshold cutoff of 20,000 that effectively separates a positive signal from a negative signal was chosen based on preliminary experiments. Negative controls of DNase/RNase free water and positive controls of known concentration E.coli bacteria were used to run alongside the samples. All samples were run in duplicates. 2.2.4 MiSeq	sequencing	pipeline       A pooled library consisting of all the samples with individually labeled indices were generated using the protocol adopted from a dual-index sequencing strategy published by Kozich et al81. As shown in Figure 2, the first step in the pipeline was to perform a touchdown PCR. For each 20ul PCR reaction, 2 ul of 10X AccuPrime PCR Buffer II, 0.15 ul of AccuPrime Taq DNA Polymerase High Fidelity, 2 ul of forward and reverse primer mixture at 0.5 uM each, 2 ul of DNA template, and 13.85 ul of distilled water were used. Each primer sequences was individually labeled with unique index sequences for identification, as shown in Table S4 and S5 (Appendix D). The cycling conditions of touchdown PCR was adapted from the protocol of Korbie & Mattick: 1 cycle at 95°C for 2 minutes, 20 cycles at 95°C for 2 minutes 60°C /54°C for 15 seconds and 72°C for 90 seconds (temperature descend at 0.3°C per cycle), 20 cycles at 95°C for 20 seconds 55°C for 15 seconds and 72°C for 90 seconds, 1 cycle at 72°C for 5 minutes82. The touchdown PCR amplified 16S rRNA gene fragments spanning the V4 region. After the PCR, products were purified with Agencourt AMPure XP system (Beckman Coulter, Catalog #A63880) following manufacturer’s protocol. Samples were randomly selected for quality check using gel electrophoresis. All tested samples had one single strong band which indicated they were 	17	qualified to be pooled. The library pool was generated using the SequalPrepTM normalization plate (Thermo Fisher Scientific, Catalog #A1051001). 25 ul of purified PCR product was transferred to the plate and sequentially eluted in 20 ul of elution buffer. DNA quality and quantity check was carried out using Agilent high sensitivity DNA kit (Agilent, Catalog #5067-4626). Lastly, sequencing was performed on the Illumina MiSeqTM platform (Illumina, Redwood City, CA, USA) with 2 x 250 paired end-read chemistry at the UBC Faculty of Pharmaceutical Science Sequencing Centre.   Figure 2. Pipeline to generate a library pool for 16S rRNA gene MiSeq sequencing        The protocol described by Kozich et al. was used for the raw sequencing data cleanup with the program Mothur (V1.34.4)81. After processing, sequence cleanup and chimera removal using the batch code (Appendix E), a total of 2,639,010 reads remained. Summary of reads for each patient can be found in Appendix E. All sequences were subsampled with 2347 sequence reads. Since there were only 812 reads for sample BIDC24, it was removed from downstream analysis. 2.2.5 Microbiome	analysis	and	statistics Two diversity metrics were assessed in the microbiome analysis. The first was alpha diversity, which was measured by Shannon diversity index, richness of the species (the number of taxa), and the other was evenness (the relative abundance of each species). The Touchdown	PCR Purification Quality	check Library	poolDNA	Quality	&	Quantity	CheckMiSeqSeqnecing	18	equation for calculating Shannon diversity index is H=EH x lnS, where H is the Shannon diversity index, EH is the evenness, and S is the richness. This was performed for each patients in R (V3.2.0) using the Vegan package (V2.3-0)83. The second diversity metric was beta diversity, which measures how different communities are structured in different patients. This is visualized by principal components analysis (PCA) of pair-wise Bray-Curtis dissimilarities and tested with permutational multivariate analysis of variance (perMANOVA) in R. If the perMANOVA value reached significance (<0.05), the Operational Taxonomic Units (OTUs) that were most likely driving the significance may be identified using Boruta feature selection after Random Forest analysis84.  2.3 Results	2.3.1 16S	rRNA	bacterial	load	analysis	      Figure 3 shows the bacterial load of each disease group measured by 16S rRNA. In the HIV-uninfected (HIV-) control cohort, there were 24 individuals with COPD (COPD+) and 24 individuals without COPD (COPD-). In the HIV-infected (HIV+) population, 7 subjects were diagnosed with COPD and 14 subjects did not have COPD. No spirometry information was recorded for the other 7 HIV+ patients, due to either loss to follow up or death.   	19	 Figure 3. 16S rRNA bacterial load across different disease status. 16S rRNA bacterial loads were 0.064 ± 0.15 copies/ng for the HIV+COPD- group; 0.83 ± 2.20 copies/ng for the HIV+COPD+ group, 2.41 ± 6.25 copies/ng for the HIV-COPD+ groups and 4.09 ± 10.25 copies/ng for the HIV-COPD- group.        According to the D’Agostino-Pearson test, data points in each group were not normally distributed. Thus, non-parametric statistical tests were applied based on distribution free methods in the following analyses. No significant differences were found across the four groups based on ANOVA test (P=0.0940). In the HIV- cohort, there was no significant difference (P=0.7186) between COPD+ and COPD- individuals. Similarly, there was no significant difference between HIV+COPD+ and HIV+COPD- patients (P=0.6533). Interestingly, when comparing the HIV+ with HIV- cohorts (0.27 ± 1.10 copies/ng in HIV+, 3.25 ± 8.44 copies/ng in HIV-), a significant difference (P=0.0076) was found [Figure 4].  16S LoadCopies/ngHIV+COPD-HIV+COPD+HIV-COPD+HIV-COPD--1001020304050	20		 Figure 4. 16S bacterial load between HIV infected and uninfected subjects  2.3.2 Phyla	distribution       Figure 5 shows the relative distribution of phyla between HIV+ patients (n= 28) and HIV- individuals (n= 48). A significant increase was observed in Proteobacteria in HIV+ patients (0.38±0.25 in HIV+, 0.19±0.19 in HIV-, P=0.0003). Decreased abundance of Bacteroidetes (0.23±0.19 in HIV+, 0.35±0.16 in HIV-, P=0.0068) and Firmicutes (0.18±0.17 in HIV+, 0.33±0.16 in HIV-, P=0.0002) were found in the HIV+ group compared to the HIV- group.  16S Load-HIVCopies/ngHIV+HIV-01020304050	21	 Figure 5. Phyla distribution between HIV positive and negative populations.        Figure 6 shows the distribution of phyla across all four groups. According to a Kruskal-Wallis Test, there was a significant difference in Bacteroidetes (P=0.0031), Proteobacteria (P=0.0013) and Firmicutes (P=0.0017) across the four groups. Post-hoc tests were performed using Dunn’s Multiple Comparison method; the differences observed in Bacteroidetes and Proteobacteria were driven by the difference between the HIV+ and HIV- groups in the COPD- population [Table S6, S7, Appendix F]. However, the difference observed in Firmicutes was driven by the difference between the HIV+ and HIV- groups in the COPD+ population [Table S8, Appendix F]. There was no significant difference between COPD+ patients and COPD- patients. 0.000.250.500.751.00HIV− HIV+Relative AbundancePhylumActinobacteriaBacteroidetesFirmicutesProteobacteriaunclassifiedOthersPhylum Distribution in Lung	22		Figure 6. Phylum distribution across all groups  2.3.3 Bacterial	diversity	analysis       Bacterial diversity was assessed by Shannon diversity, richness and evenness. As shown in Figure 7A, there was a significant difference in Shannon diversity between the HIV+ and HIV- groups (P=0.024). Shannon diversity of HIV+ patients was lower than that of HIV- individuals (1.82±0.10 for HIV+, 2.20±0.073 for HIV-). There was also a decrease in bacterial richness in HIV+ individuals (23.29±2.75 for HIV+ and 46.04±3.716 for HIV-, P<0.0001) [Figure 7B]. However, no significant difference was observed in evenness between the two groups (0.60±0.032 for HIV+, 0.61±0.017 for HIV-, P=0.51) [Figure 7C].   0.000.250.500.751.00HIV−COPD− HIV−COPD+ HIV+COPD− HIV+COPD+Relative AbundancePhylumActinobacteriaBacteroidetesFirmicutesFusobacteriaProteobacteriaVerrucomicrobiaunclassifiedOthersPhylum Distribution	23	 Figure 7. Microbial diversity between HIV+ and HIV- subjects. A) Shannon diversity index B) richness and C) evenness in HIV+ (red) and HIV- (green) subjects.         Figure 8 shows the microbial diversity in all four groups. Differences assessed by Krustal-Wallis test showed siginificance in Shannon diversity (P=0.0195) and richness (P=0.0006). Furthermore, Dunn’s Multiple Comparison Test showed that significances in diversity and richness were driven by the differences between HIV+ vs HIV- rather than the differences in COPD status [Table S9-S11, Appendix F]. Since no alpha diversity difference was observed between the COPD+ and COPD- groups, no beta diversity analysis was performed.  	Figure 8. Microbial diversity across all groups. A) Shannon diversity index B) richness and C) evenness in HIV+ (red) and HIV- (green) subjects 	24	      A rarefaction curve was plotted to assess the species richness and the optimal sampling size. As shown in Figure 9, overall, more species were found in HIV- population (blue lines) compared with HIV+ population (red lines), which agrees with the result from the Shannon diversity analysis. Most samples reached species saturation for subsampling at 3000 sequences.  Figure 9. Rarefaction curve between HIV+ (red) and HIV- (blue) subjects. Each line represents a study subject. Sample size on the x-axis means the number of sequences by random sampling. Species on the y-axis shows the number of species found by random sampling.  		25	2.3.4 Microbial	compositional	analysis       Microbial composition was analyzed using non-metric multidimensional scaling analysis as shown in Figure 10. A significant difference (perMANOVA=0.001) in bacterial community composition between HIV+ and HIV- individuals was observed.   Figure 10. Non-metric multidimensional scaling analysis that showing the bacterial compositional differences between individuals in HIV+ and HIV- groups. The blue dots represent HIV- patients and the red dots represents HIV+ patients.   	26	      Using Boruta feature selection with Random Forest analysis, six OTUs were found to be important in discriminating the HIV+ group from the HIV- group. Figure 11 is a heatmap with yellow representing a lower relative abundance, red representing a higher relative abundance and blue representing samples that did not contain such particular OTUs. OTUs that aligned to Veillonelaceae, Fusobacterium, Verrucomicrobiaceae and Campylobacter were not found in the HIV+ population. Moreover, Prevotella and Veillonella were mainly present in HIV- group. The finding suggested that these six discriminative OTUs were able to separate between the HIV-infected and uninfected population.    Figure 11. Heatmap of Boruta picked discriminative OTUs between HIV-infected and uninfected individuals 	Campylobacter	Veillonella	Fusobacterium	Verrucomicrobiaceae;	unclassified	BCCA1BCCA10BCCA11BCCA12BCCA13BCCA14BCCA15BCCA16BCCA17BCCA18BCCA19BCCA2BCCA20BCCA21BCCA22BCCA23BCCA24BCCA25BCCA26BCCA27BCCA28BCCA29BCCA3BCCA30BCCA31BCCA32BCCA33BCCA34BCCA35BCCA36BCCA37BCCA38BCCA39BCCA4BCCA40BCCA41BCCA42BCCA43BCCA44BCCA45BCCA46BCCA47BCCA48BCCA5BCCA6BCCA7BCCA8BCCA9BIDC1BIDC10BIDC14BIDC15BIDC16BIDC17BIDC18BIDC19BIDC2BIDC20BIDC21BIDC22BIDC23BIDC25BIDC26BIDC27BIDC28BIDC29BIDC3BIDC30BIDC31BIDC32BIDC33BIDC4BIDC6BIDC7BIDC8BIDC9Otu00001Otu00002Otu00012Otu00013Otu00040Otu00056Otu00767otu.wo.nc...1. otu.wo.nc...1.HIV−HIV+00.10.20.30.40.50.6Veillonellaceae;	unclassified	Prevotella	HIV+	HIV-		27	2.4 Discussion	      This study is the first to examine the lung microbiome of the small airway epithelial cells in HIV infection. We found that total bacterial 16S loads, microbial diversity and composition were significantly different between HIV+ patients and HIV- individuals. We also identified unique OTUs that could discriminate between the two groups. In contrast, the lung microbiome did not differ significantly in overall analyses between individuals with and without COPD, regardless of HIV status.       From the results of ddPCR, we found that there was no significant difference in total bacterial 16S load across all groups with respect to HIV and COPD status. Moreover, COPD status was the main contributor for the non-significant P-value. It is noteworthy that according to Mann-Whitney U test, the total bacterial 16S load in bronchial brushes from the control cohort was significantly higher than the HIV+ cohort. Although HIV-infected individuals harbor an increased susceptibility to bacterial infection and would be expected to have higher bacterial load, most HIV+ patients were on HAART and had a few prior antibiotic treatment. Such finding suggests that HAART and antibiotic treatment may play an important role in repressing microbes in the lungs. It is also important to realize that most patients had low or no bacterial 16S load detected. Although a previous study found the ddPCR 16S assay performed better compared with 16S qPCR80, the intrinsic nature of the lung, with microbes in scarce quantity in the lower respiratory tract, makes the detection of bacteria technically difficult. Nevertheless, this quantitative analysis provided a global view of the total number bacteria in the lung, which is useful for confirming and validating downstream assays.  	28	      We examined phyla distribution and found a significant phyla shift between HIV+ and HIV- groups, regardless of COPD status. For individuals infected with HIV, we found increased Proteobacteria and decreased Bacteriodetes and Firmicutes. This finding is consistent with other studies of HIV infection looking at the intestinal mucosal microbiome in HIV+ subjects85. Although those studies were performed in the context of the gut, they might provide mechanistic insights to the outgrowth of Proteobacteria in general diseased states. Studies have shown that members of the phylum Proteobacteria are linked with dysbiosis and disease86. They found a protective role of the commensal microbiota against infection or inflammation in the immune response of healthy individuals86,87. HIV infection, which disrupts homeostasis through chronic inflammation, could cause dysbiosis with a bloom of Proteobacteria in the lung. Consequently, increased Proteobacteria might further facilitate inflammation and invasion by exogenous pathogens, which in turn contribute to the progression of HIV infection.        Our diversity analysis suggests that bacterial microbiome diversity and richness from small airway epithelial cells decreased in patients infected with HIV, consistent with another study using BAL77. In fact, declines in microbiome diversity have been well characterized in many human diseases. Sze et al. had reported declines in lung microbiome diversity and richness in patients with severe COPD55. However, in our study, no significant differences were observed between patients with COPD and without COPD. It is noteworthy that most patients in our cohorts had mild to moderate COPD compared to patients with severe COPD in Sze’s cohort. The microenvironment of the lungs may be different in patients with mild COPD versus severe COPD, and this might be one factor that contributes to the discrepancy.  	29	To further explore the lung microbiome in patients infected with HIV, microbial composition analysis was performed using an NMDS plot. A significant difference (perMANOVA=0.001) was observed indicating that the composition of bacterial communities was different between HIV-infected and uninfected patients. Together with significant differences observed in alpha diversity, these findings convey a strong message that the lung bacterial microbiome is indeed alerted in patients infected with HIV.  However, a mock community with defined mixture of microbial cells was not co-sequenced with the patients’ samples. Therefore, error rates cannot be assessed. The statistical analyses would not be affected by lacking of error rate assessment since all downstream statistics were based on comparative analyses. Future experiment should include a mock community to assess the curation of sequence data.  One of potential limitations was the difficulty of data collection on HIV+ patients. The physiological information was incomplete since some patients were too ill to perform any pulmonary tests or were lost to follow up. For these reasons, the diagnosis of COPD and their severity were hard to determine. 25% of patients infected with HIV were thus removed due to lack of information. Another potential limitation of the data set was the lack of longitudinal samples and proper baseline controls. The existence of a distinct microbiome in the lower respiratory tract has been controversial. Questions have been raised whether the bacteria observed in the lungs in fact represent microaspiration of oral microbes. Previous studies have found that overall, the lung microbiome resembles that of the mouth, however, these studies had limited number of subjects 53,54,88. Nevertheless, to address these concerns, a proper oral wash would be collected as a baseline control for lung microbiome studies. This study was not originally designed for microbiome research, thus there was no oral wash 	30	collected before or after the bronchoscopy. However, emphasis on oral wash as control of lung microbiome studies was mainly for specimens collected from BAL. Cytology brush from small airway epithelium examines cells integral to the lungs, which are intrinsically different to BAL specimens, therefore should provide a more valuable perspective in residential lung microbiome. Besides, all bronchoscopies were performed with minimal suction upon entry into the lung so that contamination from the upper respiratory airway was minimized. In order to fully address the controversy, standard operating procedure (SOP) should be modified for oral brush samples instead of oral wash samples in the future.  Another potential limitation was that the composition of lung microbiome could be influenced by the location of sample collecting. In order to avoid sampling at the site of infection, bronchial brushes were not always collected at the same sites from each patient. It was shown that the distribution of inflammatory cell infiltration in the lung is not uniform throughout the bronchial tree89. This suggested the possibility of different microenvironments at various airway sites, which could be a confounder of the findings and may contribute to the significance of the microbiome analysis. However, no known study has investigated the uniformity of the lung microbiome. Future study is needed to explore the influence of the lung geometry on microbiome diversity and composition.  In summary, it is shown that the bacterial microbiome from small airway epithelial cells within the lungs changes significantly in patients infected with HIV. These changes are marked by decreased diversity and decreased richness with the abundance of certain OTUs able to distinguish HIV from non-HIV samples. In the next chapter, methylation patterns of patients infected with HIV and those uninfected were investigated to identify patterns that were specifically associated with HIV infection.  	31	3 CHAPTER	III	Methylation	in	the	Lung	of	HIV	Infected	Population 3.1 Introduction       Emerging studies have reported the involvement of epigenetic mechanism in the development of human diseases such as cancer, chronic kidney disease, mental retardation, cystic fibrosis and cardiovascular disease90–94. Epigenetic patterns of HIV infected population are distinctly different from normal population. Certain regulatory regions get either hypermethylated or hypomethylated during HIV infection, which might be associated with ageing. In fact, researchers have found genes that regulated by the change in methylation in HIV infection. For example, the cis-region of CC chemokine receptor 5 (CCR5), a critical determinant of HIV/AIDS susceptibility, is demethylated in T cells of HIV-infected individuals95. Despite a handful of studies investigating methylation pattern in HIV infection, there are no known published papers exploring methylation profiles in the lungs of HIV-infected individuals.        In this study, DNA methylation patterns from small airway epithelial cells were analyzed using bisulfite microarray technology.  The extracted DNA was treated with bisulfite to convert cytosine residues to uracil, while leaving 5-methylcytosine residues unaffected. Then, the DNA-methylation-specific mutations introduced by bisulfite treatment were mapped using a genotyping microarray. The output information provides the methylation status of DNA on a single-nucleotide resolution level.  Following the microarray, differentially methylated regions were assessed using a linear regression model. The regions were further enriched into pathways that allow for biological interpretation. By studying the differentially methylated regions between HIV-infected and uninfected populations, we 	32	anticipated unfolding the epigenetic changes that might contributed to the pathogenesis of COPD in HIV infection. 3.1 Materials	and	Methods	 3.1.1 DNA	methylation	profiling	      Unmethylated cytosine residues from the extracted DNA were converted to uracil using the EZ DNA MethylationTM Kit following the manufacturer’s protocol. All samples were then sequenced using the Illumina Infinium platform at The Centre for Applied Genomics (TCAG), The Hospital for Sick Children (SickKids) in Toronto. The HumanMethylation 450K BeadChip array was used, which interrogates 485,512 CpG sites covering 99% of RefSeq genes with an average of 17 CpG sites per gene region. It covers 96% of CpG islands, with additional coverage in islands shores and the regions flanking them. The unprocessed DNA microarray data was collected for downstream analysis as shown in Figure 12. 3.1.2 Data	processing	and	quality	control	      The Illumina Infinium data processing comprised image processing and data normalization. Microarray images were processed using Illumina BeadScan software. Signal normalization and background subtraction were performed using positive and negative control probes. β-values, which represent the absolute DNA methylation level, were calculated using ratio of intensities between methylated and unmethylated probes. Despite the used of normalization algorithms for reducing technical artifacts, batch effects are almost always present in large-scale Infinium data sets96. Thus, quality control checking probe 	33	correlation of HIV and COPD status, sex chromosomes and autosomes clustering was performed.  	Figure 12. Workflow for analyzing and interpreting DNA methylation data Bisulfite	microarray•Bisulfite	treatment	converts	cytosine	to	uracil•The	DNA-methylation-specific	mutations	are	mapped	using	microarray	chipsData	processing	and	quality	control•Microarray	data	normalzation•Quantification	of	absolute	DNA	methylation	at	single-base	resolution	(β-value)•Quality	controlStatistical	analysis•Statistically	testing	for	differentially	methylation	at	CpG	sites•Rank	CpGs	based	on	significance	and	effect	sizeInterpretation•Identifying	significant	enrichment	of	gene	functions,	regulatory	elements	and	pathyways	among	the	DMRs•Assess	and	adjust	for	confounding	factors	34	3.1.3 Statistical	analysis	      After an initial analysis of DNA methylation patterns, the next step was the identification of differentially methylated regions (DMRs) that exhibit different DNA methylation levels between HIV-infected individuals and uninfected individuals. DMR detection was carried out using β-values produced from the CpG methylation table. The Limma package (V. 3.3) in R was used to fit a linear model to the methylation data for each CpG probe97. This allows comparisons between many CpG sites simultaneously. The output was a summarized table with top ranked sites. Since a large number of CpG sites were statistically tested for differences, corrections for multiple hypothesis testing (adjusted p-values) were performed to control the false discovery rate. Statistical comparisons produced a list of CpG sites that were differentially methylated between HIV+ and HIV- groups, which provides the basis for biological interpretation downstream.  3.1.4 Interpreting	differences	in	DNA	methylation	      The DMRcate package (V. 1.8.0) in R was used to filter out probes possibly confounded by single nucleotide polymorphisms (SNPs) and cross-hybridization and to annotates gene associations98. The gene sets were then enriched and characterized through WEB-based Gene Set AnaLysis Toolkit (WebGestalt)99,100. This integrated data mining system utilized the PANTHER Classification System with GO annotation (developed by the Gene Ontology Consortium) to calculate the enrichment in biological process, molecular function and cellular component. The output table listed significantly shared GO terms, background frequency, sample frequency and P-values for biological interpretation.  		35	3.2 Results	      Among the 412,932 CpG sites listed on the output table from the limma differential analysis, there were 210,384 (50.95%) sites that had adjusted P-values less than 0.05 when comparing HIV+ to HIV- individuals. The complete result table is not shown here due to the large size. Alternatively, a panel of selected top six CpG sites with adjusted P values ranging from 4.24 × 10-48 to 7.29 × 10-46 is shown in Figure 13 for visualization.   Figure 13. β-values for top 6 CpG sites from limma result top grouped by HIV status. On each plot, the left column represents HIV- and the right represents HIV+. The green colored dots are COPD-, the blue dots are COPD+, and the pink dots are patients with unknown COPD status.  	36	      DMRs enrichment analysis identified 74 pathways in biological processes with significant P-value ranging from 2.34 × 10-9 to 0.0016. The majority of them were related to the development of nervous system, cell-cell adhesion, cell differentiation, and cell surface receptor signaling pathway [Table S12, Appendix G]. Cellular components associated with differentially methylated regions were mainly located in neurons, microtubule cytoskeleton, membrane coat, nuclear lumen and nucleoplasm [Table S13, Appendix G]. A total of 37 cellular component pathways were found with P-value ranging from 5.23 × 10-10 to 0.0056. In terms of molecular functions, 25 pathways were identified with significant P-value ranging from 2.65 × 10-12 to 0.0027. They were involved in promoter and sequence-specific DNA bindings, metal ion binding, protein binding, phospholipase activity, GTPase regulator activity, threonine-type peptidase activity and oligopeptide transporter activity [Table S14, Appendix G].   Figure 14. Principle component analysis plot with HIV status. Red represents HIV- individuals and yellow represents HIV+ individuals. 	37	 Figure 15. Principle component analysis plot with plating information.         During quality control processing, a batch effect was found in 10 HIV+ individuals as shown in Figure 14. The PCA plot was measured with 456,838 probes. PC 1 on x-axis explained 65.0%, almost 2/3, of the variations in the entire data set, whereas PC 2 on y-axis explained 6.85% of the variations. Most data points clustered on the left of the panel, where 10 HIV+ data points were clustered on the right of the panel. Surprisingly, plating was not the cause of the batch effect as shown in Figure 15. The 10 samples were on the same plate with another 3 samples that clustered with the rest of the samples.        To address the concern for batch effect, differential methylation analysis was performed twice, one with the whole data set (n63), and the other one with the 10 HIV+ samples 	38	removed (n53). For the limma results, 56.5%(192,548) of the total CpG sites overlapped in n53 and n63 dataset [Figure 16A]. 38.3% (130426) CpG sites were unique to n63 dataset and 5.2% (17836) CpG sites were unique to n53 dataset. For DMR enrichment analysis, 19.2% (1162) DMRs overlapped in n53 and n63 datasets. 79.8% (4821) DMRs were found in n63 dataset but not in n53 dataset, and only 0.9% (56) DMRs were unique in n53 dataset [Figure 16B].    Figure 16. Venn diagram of Limma and DMRcate result. A) shows the Venn diagram of Limma result. Outer circle represents CpG sites identified with the whole data set, and the inner circle represents CpG sites when the 10 HIV+ individuals were removed. Likewise, B) shows the Venn diagram of DMRs result.   	39	      Individual DMRs were inspected using a volcano plot as shown in Figure 17. The plot allows visualization of global properties in DNA methylation by showing the relationship between statistical significance on y-axis and the magnitude of differences in methylation between HIV+ and HIV- groups on x-axis. The DMRs in blue were hypermethylated in HIV+ population whereas the DMRs in red were hypomethylated.   Figure 17. Volcano plot in DNA methylation between HIV+ and HIV- subjects (n=63). 		40	3.3 Discussion	By comparing methylation patterns between patients infected with HIV and uninfected individuals, we discovered distinct pathways in biological processes, cellular components and molecular function that associated with HIV infections. For methylation patterns of biological processes, we found cell-cell adhesion and cell surface receptor signaling pathways were significantly different between the two groups. In fact, during HIV infection, binding and entry of the virus into the host cell plays a major role in determining the ability of HIV to injure the human immune system. Our finding suggested that adhesion of virus to the host cells and receptor-assisted fusion of the cell and viral membranes may be crucial in the replication cycle of HIV. In addition, we also found that methylation patterns were different in genes involved in cell differentiation between patients infected with HIV and those who were uninfected. Not surprisingly, HIV infection has been linked with massive activation and differentiation of T-cells, which is consistent with our findings.  Two main cellular components from the intracellular compartment were differentially methylated between HIV+ and HIV- individuals. Microtubule cytoskeleton was identified as one of these cellular components. Interestingly, a link between microtubule polymerization and the pro-apoptotic effect of HIV-encoded proteins has been discovered101. Studies have found that the protein released by HIV-infected cells could cause changes in mitochondrial membrane permeability, which further interfere with the microtubule polymerization102–104. Mitochondria are strongly associated with oxidative stress since they are involved in the production of reactive oxygen species in the respiratory chain105. Interestingly, altered oxidant/antioxidant level plays a key role in the vicious cycle hypothesis [Figure 1], which 	41	might be linked to the observation of altered microbiome in HIV-infected patients. Another cellular component was nucleoplasm. Since HIV replicates through integration of its DNA into the host DNA, it is expected to detect differentially methylated patterns within nucleus. Analyzing methylation patterns in terms of cellular components may shed light for future drug targets to prevent the lung complications associated with HIV.  One limitation of this study was the batch effect. 10 patients infected with HIV had a distinct methylation pattern compared with the rest of samples. Surprisingly, the batch effect was not caused by plating. Since the 10 samples were extracted at an earlier time, we suspected that the batch effect was caused by the variation in DNA stability. In order to achieve optimal bisulfite sequencing result, the loss of DNA due to non-specific degradation should be kept to a minimum as much as possible106. DNA stored for a longer time may be more prone to DNA degradation during bisulfite conversion. Thus, bisulfite conversion for the 10 samples may not have been as efficient as the rest of the sample set. To address this limitation, both n63 and n53 data set were analyzed. 56.5% of the total CpG sites overlapped between n53 and n63 datasets, and only 19.2% of DMRs overlapped between n53 and n63 datasets. Therefore, our result using n63 dataset may contain false positive signals that caused by the batch effect, whereas n53 dataset may miss some true positive signals. For future experiment, all samples should be extracted and processed at the same time to avoid experimental variations caused by execution time.  To conclude, we discovered significant differences in methylation patterns between patients infected with HIV and patients without HIV. Although a batch effect was observed in the 10 samples from HIV+ cohort, a significant difference was still observed after removing the 10 samples. This study identified specific pathways that involved in the 	42	differentially methylated regions within the lungs. These findings provide promising opportunities to target treatment for lung diseases caused by HIV infection in the future. 	43	CHAPTER	IV	CONCLUSION This research showed that the small airway epithelial cell bacterial microbiome and host methylation patterns are different within the lungs during HIV infection. In Chapter 2, decreased microbial diversity and richness were found in patients infected with HIV. Certain OTUs were identified to distinguish patients with HIV infection from those uninfected individuals. Characterization of the lung microbiome has shed light on our understanding of the complex microbial ecosystem and its role in HIV associated COPD. In Chapter 3, methylation profiles of patients infected with HIV and uninfected population were investigated. A significant difference in global methylation patterns with identified CpG sites and differentially methylated regions were reported. The finding provides a good first step for future investigation on how methylation patterns might affect gene expression and disease progression. Future studies based on larger sample sizes with collection of proper oral wash controls and experimental consistency will be able to overcome the limitations and improve our understanding of the role of lung microbiome in HIV infection as well as how the virus might influence the host methylation profiles. Although the research into the lung microbiome is still in its infancy, investigation on how lung microbiome imbalance might influence disease progression and the effect of the restoration of homeostasis need to be studied. Moreover, in order to understand the mechanism on how the lung microbiome might contribute to the disease progression, the associations between the lung microbiome and host methylation with gene expression need to be investigated. Overall, the work provided in this thesis explores the lung microbiome and methylation patterns between HIV-infected and 	44	uninfected individuals, which add value to the current understanding of lung complications in HIV infection.    	45	Bibliography	1.  UNAIDS. How AIDS Changed Everything MDG 6: 15 Years, 15 Lessons of Hope from the AIDS Response.; 2015. 2.  UNAIDS. Fact sheet: Global HIV/AIDS Statistics. UNAIDS. Retrieved May 3rd, 2016 from http://www.unaids.org/en/resources/campaigns/2014/2014gapreport/factsheet. Published 2015. 3.  Gottlieb MS, Schroff R, Schanker HM, et al. Pneumocystis carinii Pneumonia and Mucosal Candidiasis in Previously Healthy Homosexual Men. N Engl J Med. 1981;305(24):1425-1431.  4.  Barre-Sinoussi F, Ross AL, Delfraissy J-F. Past, present and future: 30 years of HIV research. Nat Rev Micro. 2013;11(12):877-883.  5.  Popovic M, Sarngadharan M, Read E, Gallo R. Detection, isolation, and continuous production of cytopathic retroviruses (HTLV-III) from patients with AIDS and pre-AIDS. Science (80- ). 1984;224(4648):497-500.  6.  Levy J a, Hoffman  a D, Kramer SM, Landis J a, Shimabukuro JM, Oshiro LS. Isolation of lymphocytopathic retroviruses from San Francisco patients with AIDS. Science. 1984;225(4664):840-842. 7.  Dalgleish AG, Beverley PC, Clapham PR, Crawford DH, Greaves MF, Weiss RA. The CD4 (T4) antigen is an essential component of the receptor for the AIDS retrovirus. Nature. 1984;312(5996):763-767.  8.  Klatzmann D, Champagne E, Chamaret S, et al. T-lymphocyte T4 molecule behaves as the receptor for human retrovirus LAV. Nature. 1984;312(5996):767-768.  9.  Wain-Hobson S, Sonigo P, Danos O, Cole S, Alizon M. Nucleotide sequence of the AIDS virus, LAV. Cell. 1985;40(1):9-17.  10.  Sanchez-Pescador R, Power MD, Barr PJ, et al. Nucleotide sequence and expression of an AIDS-associated retrovirus (ARV-2). Science. 1985;227(4686):484-492.  11.  Hammer SM, Squires KE, Hughes MD, et al. A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less. N Engl J Med. 1997;337(11):725-733.  12.  Fleishman JA, Hellinger FH. Recent trends in HIV-related inpatient admissions 1996-2000: a 7-state study. J Acquir Immune Defic Syndr. 2003;34(1):102-110.  13.  Samji H, Cescon A, Hogg RS, et al. Closing the gap: Increases in life expectancy among treated HIV-positive individuals in the United States and Canada. PLoS One. 2013;8(12).  14.  Deeks SG, Lewin SR, Havlir D V. The end of AIDS: HIV infection as a chronic disease. Lancet. 2013;382(9903):1525-1533.  15.  Holmberg SD, Moorman AC, Williamson JM, et al. Protease inhibitors and cardiovascular outcomes in patients with HIV-1. Lancet. 2002;360(9347):1747-1748.  16.  Palella FJ, Baker RK, Moorman AC, et al. Mortality in the highly active antiretroviral therapy 	46	era: changing causes of death and disease in the HIV outpatient study. J Acquir Immune Defic Syndr. 2006;43(1):27-34. 17.  Wallace JM, Hansen NI, Lavange L, et al. Respiratory disease trends in the Pulmonary Complications of HIV Infection Study cohort. Pulmonary Complications of HIV Infection Study Group. Am J Respir Crit Care Med. 1997;155(1):72-80.  18.  Palella FJ, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med. 1998;338(13):853-860.  19.  Collini P, Morris A. Maintaining lung health with longstanding HIV. Curr Opin Infect Dis. 2015:31-38.  20.  Crothers K, Butt AA, Gibert CL, Rodriguez-Barradas MC, Crystal S, Justice AC. Increased COPD among HIV-positive compared to HIV-negative veterans. Chest. 2006;130(5):1326-1333.  21.  Schwarcz SK, Vu A, Hsu LC, Hessol NA. Changes in Causes of Death Among Persons with AIDS: San Francisco, California, 1996-2011. AIDS Patient Care STDS. 2014;28(10):517-523.  22.  Pefura-Yone EW, Fodjeu G, Kengne AP, Roche N, Kuaban C. Prevalence and determinants of chronic obstructive pulmonary disease in HIV infected patients in an African country with low level of tobacco smoking. Respir Med. 2015;109(2):247-254.  23.  George MP, Kannass M, Huang L, Sciurba FC, Morris A. Respiratory symptoms and airway obstruction in HIV-infected subjects in the HAART era. PLoS One. 2009;4(7).  24.  Madeddu G, Fois AG, Calia GM, et al. Chronic obstructive pulmonary disease: An emerging comorbidity in HIV-infected patients in the HAART era? Infection. 2013;41(2):347-353.  25.  Hirani A, Cavallazzi R, Vasu T, et al. Prevalence of obstructive lung disease in HIV population: A cross sectional study. Respir Med. 2011;105(11):1655-1661.  26.  Sampériz G, Guerrero D, López M, et al. Prevalence of and risk factors for pulmonary abnormalities in HIV-infected patients treated with antiretroviral therapy. HIV Med. 2013.  27.  Guaraldi G, Besutti G, Scaglioni R, et al. The Burden of image based emphysema and bronchiolitis in HIV-infected individuals on antiretroviral therapy. PLoS One. 2014;9(10).  28.  Gritz ER, Vidrine DJ, Lazev AB, Amick BC, Arduino RC. Smoking behavior in a low-income multiethnic HIV/AIDS population. Nicotine Tob Res. 2004;6(1):71-77.  29.  Niaura R, Shadel WG, Morrow K, Tashima K, Flanigan T, Abrams DB. Human immunodeficiency virus infection, AIDS, and smoking cessation: the time is now. Clin Infect Dis. 2000;31(3):808-812.  30.  Justice AC, Lasky E, McGinnis KA, et al. Medical Disease and Alcohol Use Among Veterans With Human Immunodeficiency Infection. Med Care. 2006;44(8):52-60.  31.  Morris A, George MP, Crothers K, et al. HIV and chronic obstructive pulmonary disease: is it worse and why? Proc Am Thorac Soc. 2011;8(3):320-325.  32.  Morris A, Sciurba FC, Lebedeva IP, et al. Association of chronic obstructive pulmonary 	47	disease severity and Pneumocystis colonization. Am J Respir Crit Care Med. 2004;170(4):408-413.  33.  Treitinger  a, Spada C, Verdi JC, et al. Decreased antioxidant defence in individuals infected by the human immunodeficiency virus. Eur J Clin Invest. 2000;30(5):454-459.  34.  Brinkman K, Smeitink JA, Romijn JA, Reiss P. Mitochondrial toxicity induced by nucleoside-analogue reverse-transcriptase inhibitors is a key factor in the pathogenesis of antiretroviral-therapy-related lipodystrophy. Lancet. 1999;354(9184):1112-1115.  35.  Cadenas E, Davies KJ. Mitochondrial free radical generation, oxidative stress, and aging. Free Radic Biol Med. 2000;29(3-4):222-230.  36.  Kirkham PA, Barnes PJ. Oxidative stress in COPD. Chest. 2013;144(1):266-273.  37.  Twigg HL, Soliman DM, Day RB, et al. Lymphocytic alveolitis, bronchoalveolar lavage viral load, and outcome in human immunodeficiency virus infection. Am J Respir Crit Care Med. 1999;159(5 I):1439-1444.  38.  Wang Z, Zheng T, Zhu Z, et al. Interferon gamma induction of pulmonary emphysema in the adult murine lung. J Exp Med. 2000;192(11):1587-1600.  39.  Kelly JR, Kennedy PJ, Cryan JF, Dinan TG, Clarke G, Hyland NP. Breaking down the barriers: the gut microbiome, intestinal permeability and stress-related psychiatric disorders. Front Cell Neurosci. 2015;9:392. 40.  Suau A, Bonnet R, Sutren M, et al. Direct analysis of genes encoding 16S rRNA from complex communities reveals many novel molecular species within the human gut. Appl Env Microbiol. 1999;65(11):4799-4807.  41.  Morgan XC, Tickle TL, Sokol H, et al. Dysfunction of the intestinal microbiome in inflammatory bowel disease and treatment. Genome Biol. 2012;13(9):R79.  42.  Giongo A, Gano KA, Crabb DB, et al. Toward defining the autoimmune microbiome for type 1 diabetes. ISME J. 2011;5(1):82-91.  43.  Larsen N, Vogensen FK, Van Den Berg FWJ, et al. Gut microbiota in human adults with type 2 diabetes differs from non-diabetic adults. PLoS One. 2010;5(2).  44.  Grice E a, Segre J a. The skin microbiome. Nat Rev Microbiol. 2011;9(4):244-253.  45.  Schulz MD, Atay Ç, Heringer J, et al. High-fat-diet-mediated dysbiosis promotes intestinal carcinogenesis independently of obesity TL  - 514. Nature. 2014;514 VN  - (7523):508-512.  46.  Cryan JF, Dinan TG. Mind-altering microorganisms: the impact of the gut microbiota on brain and behaviour. Nat Rev Neurosci. 2012;13(10):701-712.  47.  Jiang W, Lederman MM, Hunt P, et al. Plasma levels of bacterial DNA correlate with immune activation and the magnitude of immune restoration in persons with antiretroviral-treated HIV infection. J Infect Dis. 2009;199(8):1177-1185.  48.  Brenchley JM, Price DA, Schacker TW, et al. Microbial translocation is a cause of systemic immune activation in chronic HIV infection. Nat Med. 2006;12(12):1365-1371.  	48	49.  Hilty M, Burke C, Pedro H, et al. Disordered microbial communities in asthmatic airways. PLoS One. 2010;5(1).  50.  Dickson RP, Huffnagle GB. The Lung Microbiome: New Principles for Respiratory Bacteriology in Health and Disease. PLOS Pathog. 2015;11(7):e1004923.  51.  Dickson RP, Erb-Downward JR, Huffnagle GB. The role of the bacterial microbiome in lung disease. Expert Rev Respir Med. 2013;7(3):245-257.  52.  Charlson ES, Bittinger K, Chen J, et al. Assessing Bacterial Populations in the Lung by Replicate Analysis of Samples from the Upper and Lower Respiratory Tracts. PLoS One. 2012;7(9).  53.  Charlson ES, Bittinger K, Haas AR, et al. Topographical continuity of bacterial populations in the healthy human respiratory tract. Am J Respir Crit Care Med. 2011;184(8):957-963.  54.  Erb-Downward JR, Thompson DL, Han MK, et al. Analysis of the lung microbiome in the “healthy” smoker and in COPD. PLoS One. 2011;6(2).  55.  Sze MA, Dimitriu PA, Hayashi S, et al. The lung tissue microbiome in chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2012;185(10):1073-1080.  56.  Goldberg AD, Allis CD, Bernstein E. Epigenetics: A Landscape Takes Shape. Cell. 2007;128(4):635-638.  57.  Takahashi K, Sugi Y, Nakano K, et al. Epigenetic control of host gene by commensal bacteria in large intestinal epithelial cells. J Biol Chem. 2011;286(41):35755-35762.  58.  Olszak T, An D, Zeissig S, et al. Microbial exposure during early life has persistent effects on natural killer T cell function. Science. 2012;336(6080):489-493.  59.  Ehrlich M, Gama-Sosa MA, Huang LH, et al. Amount and distribution of 5-methylcytosine in human DNA from different types of tissues or cells. Nucleic Acids Res. 1982;10(8):2709-2721.  60.  Christensen BC, Houseman EA, Marsit CJ, et al. Aging and environmental exposures alter tissue-specific DNA methylation dependent upon CPG island context. PLoS Genet. 2009;5(8).  61.  Bollati V, Schwartz J, Wright R, et al. Decline in genomic DNA methylation through aging in a cohort of elderly subjects. Mech Ageing Dev. 2009;130(4):234-239.  62.  Teschendorff AE, Menon U, Gentry-Maharaj A, et al. Age-dependent DNA methylation of genes that are suppressed in stem cells is a hallmark of cancer. Genome Res. 2010;20(4):440-446.  63.  Bell JT, Tsai PC, Yang TP, et al. Epigenome-wide scans identify differentially methylated regions for age and age-related phenotypes in a healthy ageing population. PLoS Genet. 2012;8(4).  64.  Horvath S, Zhang Y, Langfelder P, et al. Aging effects on DNA methylation modules in human brain and blood tissue. Genome Biol. 2012;13(10):R97.  65.  Horvath S. DNA methylation age of human tissues and cell types. Genome Biol. 2013;14(10):R115.  	49	66.  Horvath S, Levine AJ. HIV-1 Infection Accelerates Age According to the Epigenetic Clock. J Infect Dis. 2015;212(10):1563-1573.  67.  Gross AM, Jaeger PA, Kreisberg JF, et al. Methylome-wide Analysis of Chronic HIV Infection Reveals Five-Year Increase in Biological Age and Epigenetic Targeting of HLA Article Methylome-wide Analysis of Chronic HIV Infection Reveals Five-Year Increase in Biological Age and Epigenetic Targeting of. Mol Cell. 2016;62(2):157-168.  68.  Sze MA, Dimitriu PA, Suzuki M, et al. Host response to the lung microbiome in chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2015;192(4):438-445. 69.  Cabrera-Rubio R, Garcia-Núñez M, Setó L, et al. Microbiome diversity in the bronchial tract of patients with chronic obstructive pulmonary disease. J Clin Microbiol. 2012.  70.  Vyboh K, Jenabian MA, Mehraj V, Routy JP. HIV and the gut microbiota, partners in crime: Breaking the vicious cycle to unearth new therapeutic targets. J Immunol Res. 2015;2015.  71.  Mammen MJ, Sethi S. COPD and the microbiome. Respirology. 2016:n/a - n/a.  72.  Sethi S, Murphy TF. Infection in the pathogenesis and course of chronic obstructive pulmonary disease. N Engl J Med. 2008;359(22):2355-2365.  73.  Murray MA, Chotirmall SH. The Impact of Immunosenescence on Pulmonary Disease. Mediators Inflamm. 2015;2015:692546.  74.  Sze MA, Dimitriu PA, Hayashi S, et al. The lung tissue microbiome in chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2012;185(10):1073-1080.  75.  Lozupone C, Cota-Gomez A, Palmer BE, et al. Widespread colonization of the lung by Tropheryma whipplei in HIV infection. Am J Respir Crit Care Med. 2013;187(10):1110-1117.  76.  Iwai S, Huang D, Fong S, et al. The lung microbiome of Ugandan HIV-infected pneumonia patients is compositionally and functionally distinct from that of San Franciscan patients. PLoS One. 2014;9(4).  77.  Twigg III HL, Knox KS, Zhou J, et al. Effect of Advanced HIV Infection on the Respiratory Microbiome. Am J Respir Crit Care Med. February 2016.  78.  Beck JM, Schloss PD, Venkataraman A, et al. Multicenter comparison of lung and oral microbiomes of HIV-infected and HIV-uninfected individuals. Am J Respir Crit Care Med. 2015;192(11):1335-1344.  79.  Rabe KF, Hurd S, Anzueto A, et al. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med. 2007;176(6):532-555.  80.  Sze MA, Abbasi M, Hogg JC, Sin DD. A comparison between droplet digital and quantitative PCR in the analysis of bacterial 16S load in lung tissue samples from control and COPD GOLD 2. PLoS One. 2014.  81.  Kozich JJ, Westcott SL, Baxter NT, Highlander SK, Schloss PD. Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the miseq illumina sequencing platform. Appl Environ Microbiol. 2013;79(17):5112-5120.  	50	82.  Korbie DJ, Mattick JS. Touchdown PCR for increased specificity and sensitivity in PCR amplification. Nat Protoc. 2008;3(9):13-15.  83.  Dixon P. VEGAN, a package of R functions for community ecology. J Veg Sci. 2003;14:927-930.  84.  Kursa MB, Rudnicki WR. Feature Selection with the Boruta Package. J Stat Softw. 2010;36(11):1-13. doi:Vol. 36, Issue 11, Sep 2010. 85.  Dillon SM, Lee EJ, Kotter C V, et al. An altered intestinal mucosal microbiome in HIV-1 infection is associated with mucosal and systemic immune activation and endotoxemia. Mucosal Immunol. 2014.  86.  Shin NR, Whon TW, Bae JW. Proteobacteria: Microbial signature of dysbiosis in gut microbiota. Trends Biotechnol. 2015;33(9).  87.  Mazmanian SK, Round JL, Kasper DL. A microbial symbiosis factor prevents intestinal inflammatory disease. Nature. 2008;453(7195):620-625.  88.  Morris A, Beck JM, Schloss PD, et al. Comparison of the respiratory microbiome in healthy nonsmokers and smokers. Am J Respir Crit Care Med. 2013;187(10):1067-1075.  89.  Olloquequi J, Ferrer J, Montes JF, Rodríguez E, Montero MA, García-Valero J. Differential lymphocyte infiltration in small airways and lung parenchyma in COPD patients. Respir Med. 2010;104(9):1310-1318.  90.  Esteller M, Herman JG. Cancer as an epigenetic disease: DNA methylation and chromatin alterations in human tumours. J Pathol. 2002;196(1):1-7.  91.  Smyth LJ, McKay GJ, Maxwell AP, McKnight AJ. DNA hypermethylation and DNA hypomethylation is present at different loci in chronic kidney disease. Epigenetics. 2014;9(3):366-376.  92.  Robertson KD. DNA methylation and human disease. Nat Rev Genet. 2005;6(8):597-610.  93.  Sirinupong N, Yang Z. Epigenetics in Cystic Fibrosis: Epigenetic Targeting of a Genetic Disease. Curr Drug Targets. 2015:976-987. 94.  Webster ALH, Yan MSC, Marsden PA. Epigenetics and Cardiovascular Disease. Can J Cardiol. 2013;29(1):46-57.  95.  Gornalusse GG, Mummidi S, Gaitan AA, et al. Epigenetic mechanisms, T-cell activation, and CCR5 genetics interact to regulate T-cell expression of CCR5, the major HIV-1 coreceptor. Proc Natl Acad Sci U S A. 2015;112(34):E4762-E4771.  96.  Teschendorff AE, Zhuang J, Widschwendter M. Independent surrogate variable analysis to deconvolve confounding factors in large-scale microarray profiling studies. Bioinformatics. 2011;27(11):1496-1505.  97.  Ritchie ME, Phipson B, Wu D, et al. Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7):e47.  98.  Peters TJ, Buckley MJ, Statham AL, et al. De novo identification of differentially methylated regions in the human genome. Epigenetics Chromatin. 2015;8(1):6.  	51	99.  Zhang B, Kirov S, Snoddy J. WebGestalt: An integrated system for exploring gene sets in various biological contexts. Nucleic Acids Res. 2005;33(SUPPL. 2).  100.  Wang J, Duncan D, Shi Z, Zhang B. WEB-based GEne SeT AnaLysis Toolkit (WebGestalt): update 2013. Nucleic Acids Res. 2013;41(Web Server issue).  101.  Chen D, Wang M, Zhou S, Zhou Q. HIV-1 Tat targets microtubules to induce apoptosis, a process promoted by the pro-apoptotic Bcl-2 relative Bim. EMBO J. 2002;21(24):6801-6810.  102.  Macho  a, Calzado M a, Jiménez-Reina L, Ceballos E, León J, Muñoz E. Susceptibility of HIV-1-TAT transfected cells to undergo apoptosis. Biochemical mechanisms. Oncogene. 1999;18:7543-7551.  103.  Ferri KF, Jacotot E, Blanco J, Esté J a, Kroemer G. Mitochondrial control of cell death induced by HIV-1-encoded proteins. Ann N Y Acad Sci. 2000;926:149-164. 104.  Battaglia P a, Zito S, Macchini  a, Gigliani F. A Drosophila model of HIV-Tat-related pathogenicity. J Cell Sci. 2001;114(Pt 15):2787-2794.  105.  Lenaz G. Role of mitochondria in oxidative stress and ageing. Biochim Biophys Acta - Bioenerg. 1998;1366(1-2):53-67.  106.  Grunau C, Clark SJ, Rosenthal  a. Bisulfite genomic sequencing: systematic investigation of critical experimental parameters. Nucleic Acids Res. 2001;29(13):E65-E65.    	52	Appendices		Appendix	A:	An	overview	of	clinical	traits	of	HIV	infected	and	uninfected	subjects		Table 2.Demographics of patients infected with HIV Age	 Sex	 Vital	Status	Current	VL	Current	CD4	Bronchoscopy	Indication	Smoking	Status	 Pack-Years	Current	HAART	CT	Emphysema	 FEV1	(L)	FEV1/	FVC	(%)	69	 M	 Alive	 <40	 440	 Cancer	 Current	 30	 Yes	 Yes	 1.53	 33.04	57	 F	 Alive	 2375	 160	 Pneumonia	 Current	 39	 No	 Yes	 1.11	 90.00	75	 M	 Alive	 <40	 590	 Cancer	 Current	 130	 Yes	 Yes	 2.99	 70.55	54	 M	 Alive	 <40	 690	 Cancer	 Current	 12.5	 Yes	 Yes	 3.35	 56.63	58	 F	 Alive	 65	 580	 Cancer	 Current	 37.5	 Yes	 Yes	 2.71	 64.18	71	 M	 Alive	 63918	 230	 Cancer	 Current	 30	 No	 No	 N/A	 N/A	50	 M	 Alive	 <40	 910	 Cancer	 Current	 15	 Yes	 No	 3.33	 70.95	55	 M	 Alive	 45877	 210	 Pneumonia	 Current	 19.5	 No	 Yes	 N/A	 N/A	63	 M	 Alive	 <40	 780	 Cancer	 Past	 20	 Yes	 Yes	 2.54	 59.86	67	 M	 Alive	 <40	 840	 Cancer	 Past	 3	 Yes	 Yes	 2.87	 71.44	67	 M	 Alive	 112	 280	 Cancer	 Past	 45	 Yes	 No	 2.41	 76.80	61	 M	 Alive	 <40	 1120	 Bronchiectasis	 Past	 12	 Yes	 No	 3.06	 70.57	62	 M	 Alive	 <40	 380	 Pneumonia	 Past	 75	 Yes	 No	 2.47	 85	46	 M	 Alive	 <40	 110	 Cancer	 Current	 30	 Yes	 Yes	 2.41	 51.56	43	 F	 Alive	 14875	 190	 Pneumonia	 Current	 115	 Yes	 Yes	 2	 78.28	56	 M	 Alive	 <40	 420	 Cancer	 Past	 90	 Yes	 Yes	 3.33	 75.76	75	 M	 Deceased	 <40	 460	 Cancer	 Past	 20	 Yes	 Yes	 2.75	 69.09	60	 M	 Deceased	 <40	 140	 Cancer	 Past	 4	 Yes	 No	 2.45	 72.86	62	 M	 Deceased	 <40	 180	 Cancer	 None	 0	 Yes	 Yes	 N/A	 N/A	51	 M	 Deceased	 <40	 150	 Pneumonia	 Current	 N/A	 Yes	 Yes	 N/A	 N/A	41	 F	 Deceased	 2430000	 60	 Cancer	 None	 0	 No	 Yes	 N/A	 N/A			 		53		Table 3. Demographics of patients in the control cohort Age	 Sex	 Smoking	Status	 Pack-Years	FEV1	(L)	 FEV1/FVC	(%)	 Cancer	Status	82	 M	 Never	 N/A	 1.6	 N/A	 CANCER	67	 F	 Past	 61.5	 0.95	 54	 N/A	63	 M	 Current	 54.6	 N/A	 51	 N/A	59	 M	 Current	 66.75	 N/A	 59	 N/A	66	 M	 Past	 41	 N/A	 40	 N/A	62	 M	 Past	 72	 1.7	 46	 N/A	75	 F	 Current	 58	 N/A	 55	 N/A	68	 M	 Current	 53	 N/A	 46	 N/A	57	 M	 Past	 40	 N/A	 36	 N/A	63	 M	 Past	 32	 N/A	 63	 N/A	77	 M	 Past	 90	 N/A	 47	 N/A	54	 M	 Current	 37.5	 N/A	 60	 N/A	64	 M	 Past	 33	 N/A	 62	 N/A	63	 M	 Current	 54	 N/A	 50	 N/A	68	 F	 Past	 63.69	 N/A	 56	 N/A	67	 F	 Current	 72	 N/A	 63	 N/A	69	 M	 Current	 65	 N/A	 63	 N/A	66	 M	 Past	 31	 N/A	 63	 N/A	57	 F	 Current	 42	 N/A	 57	 N/A	78	 F	 Past	 41	 N/A	 55	 N/A	64	 F	 Current	 46	 N/A	 54	 N/A	63	 M	 Past	 60.9	 N/A	 56	 N/A	61	 M	 Past	 42	 N/A	 56	 N/A	65	 M	 Current	 44	 N/A	 59	 N/A	48	 M	 Past	 N/A	 3.06	 N/A	 CANCER	62	 F	 Current	 N/A	 2.31	 N/A	 CANCER	49	 F	 Never	 N/A	 2.38	 N/A	 CANCER	75	 M	 Past	 N/A	 2.6	 N/A	 CANCER	68	 F	 Current	 N/A	 1.74	 N/A	 CANCER	64	 M	 Never	 N/A	 3.53	 N/A	 CANCER	60	 M	 Past	 39.09	 N/A	 81	 N/A	49	 F	 Pasr	 42	 N/A	 85	 N/A	53	 M	 Current	 34	 N/A	 80	 N/A	71	 F	 Past	 44	 N/A	 83	 N/A	54	 F	 Past	 52	 N/A	 81	 N/A	66	 F	 Past	 40	 N/A	 93	 N/A	63	 F	 Past	 45	 2.56	 80	 N/A		54	Table 3. Demographics of patients in the control cohort Age	 Sex	 Smoking	Status	 Pack-Years	FEV1	(L)	 FEV1/FVC	(%)	 Cancer	Status	53	 F	 Current	 44	 N/A	 83	 N/A	55	 F	 Past	 34	 N/A	 80	 N/A	60	 F	 Current	 38	 N/A	 82	 N/A	57	 F	 Current	 30	 N/A	 84	 N/A	56	 F	 Current	 36	 N/A	 80	 N/A	65	 M	 Past	 51	 N/A	 80	 N/A	75	 F	 Past	 75	 N/A	 84	 N/A	57	 F	 Current	 38	 N/A	 81	 N/A	58	 F	 Past	 47	 N/A	 82	 N/A	61	 M	 Current	 52	 N/A	 83	 N/A	62	 M	 Current	 41	 N/A	 81	 N/A			 		55	Appendix	B:	DNA	extraction	protocol	for	bronchial	brushing	1. Thaw	samples	on	ice.	2. Centrifuge	at	13,000	rpm	at	4°C	for	5	min	to	spin	down	the	cells.		3. Remove	supernatant	carefully.		*	Cells	are	collected	mainly	on	the	wall	of	the	tubes	or	on	the	brush.	If	cell	pellet	is	visible,	remove	as	much	supernatant	as	possible;	if	there	is	nor	visible	cell	pellet,	carefully	remove	supernatant	without	touching	the	side	wall,	leave	around	50	ul	on	the	bottom	of	the	tube.	Follow	steps	outlined	in	the	protocol	of	QIAGEN	DNeasy	Blood	&	Tissue	Kit	(Cat	No.	69504)	4. Add	20	ul	of	proteinase	K	and	180	ul	of	ATL.	Mix	by	pulse	vortexing	for	15	second.	5. Incubate	samples	at	56°C	until	completely	lysed.	Vortex	occasionally	during	incubation.	*	Incubate	at	least	30	min.	Samples	can	be	left	at	56°C	overnight.			6. Pulse	vortex	tubes	for	15	second	and	centrifuge	briefly.	7. Add	200	ul	buffer	AL.	Mix	thoroughly	by	vortexing.	8. Add	200	ul	freshly	made	ethanol	(96-100%).	Mix	thoroughly	by	vortexing.	9. Load	the	mixture	into	a	DNeasy	Mini	spin	column	placed	in	a	2	ml	collection	tube.	Centrifuge	at	≥	6,000	x	g	(8,000	rpm)	for	1	min.	Discard	the	flow-through	and	collection	tube.	10. Place	the	spin	column	in	a	new	2	ml	collection	tube,	add	500	ul	Buffer	AW1.	Centrifuge	for	1	min	at	≥	6,000	x	g	(8,000	rpm)	for	1	min.	Discard	the	flow-through	and	collection	tube.	11. Place	the	spin	column	in	a	new	2	ml	collection	tube,	add	500	ul	Buffer	AW2.	Centrifuge	for	5	min	at	20,000	x	g	(14,000	rpm)	for	1	min.	Discard	the	flow-through	and	collection	tube.	12. Transfer	the	spin	column	to	a	new	1.5	ml	microcentrifuge	tube.	13. Elute	the	DNA	by	adding	50	ul	distilled	water	to	the	center	of	the	spin	column	membrane.	Incubate	for	1	min	at	room	temperature.	Centrifuge	for	1	min	at	≥	6,000	x	g	(8,000	rpm).	14. Optional:	Repeat	step	13	for	increased	DNA	yield.	15. Nanodrop	to	measure	the	concentration.	If	not	used	within	two	weeks,	store	at	-80°C.		 		56	Appendix	C:	DDPCR	primer	sequences	Table 4. Forward and reverse primer sequences for 16S rRNA gene used for ddPCR Target	 Primer	Name	 Primer	Sequence	16S	rRNA	Gene	 63F	 5’-	GCAGGCCTAACACATGCAAGTC-3’	355R	 5’-	CTGCTGCCTCCCGTAGGAGT-3’				 		57	Appendix	D:	Touchdown	PCR	primer	sequences	for	16S	rRNA	gene	MiSeq	sequencing		Each	primer	consists	of	the	Illumina	adapter	(in	green),	an	8-nt	index	sequence	(in	red),	a	10-nt	pad	sequence	(in	purple),	a	2-nt	linker	(in	yellow),	and	the	gene-specific	primer	(in	blue).	The	complete	primers	were	each	64	to	68	bp	long.			Table 5. Forward primer sequences for 16S rRNA gene used for MiSeq sequencing Primer	ID	 Forward	Full	Sequence	v4.SA501	 AATGATACGGCGACCACCGAGATCTACACATCGTACGTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SA502	 AATGATACGGCGACCACCGAGATCTACACACTATCTGTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SA503	 AATGATACGGCGACCACCGAGATCTACACTAGCGAGTTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SA504	 AATGATACGGCGACCACCGAGATCTACACCTGCGTGTTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SA505	 AATGATACGGCGACCACCGAGATCTACACTCATCGAGTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SA506	 AATGATACGGCGACCACCGAGATCTACACCGTGAGTGTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SA507	 AATGATACGGCGACCACCGAGATCTACACGGATATCTTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SA508	 AATGATACGGCGACCACCGAGATCTACACGACACCGTTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SB501	 AATGATACGGCGACCACCGAGATCTACACCTACTATATATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SB502	 AATGATACGGCGACCACCGAGATCTACACCGTTACTATATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SB503	 AATGATACGGCGACCACCGAGATCTACACAGAGTCACTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SB504	 AATGATACGGCGACCACCGAGATCTACACTACGAGACTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SB505	 AATGATACGGCGACCACCGAGATCTACACACGTCTCGTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SB506	 AATGATACGGCGACCACCGAGATCTACACTCGACGAGTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SB507	 AATGATACGGCGACCACCGAGATCTACACGATCGTGTTATGGTAATTGTGTGCCAGCMGCCGCGGTAA	v4.SB508	 AATGATACGGCGACCACCGAGATCTACACGTCAGATATATGGTAATTGTGTGCCAGCMGCCGCGGTAA									58		Table 6. Reverse primer sequences for 16S rRNA gene used for MiSeq sequencing Primer	ID	 Reverse	Full	Sequence	v4.SA701	 CAAGCAGAAGACGGCATACGAGATAACTCTCGAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SA702	 CAAGCAGAAGACGGCATACGAGATACTATGTCAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SA703	 CAAGCAGAAGACGGCATACGAGATAGTAGCGTAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SA704	 CAAGCAGAAGACGGCATACGAGATCAGTGAGTAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SA705	 CAAGCAGAAGACGGCATACGAGATCGTACTCAAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SA706	 CAAGCAGAAGACGGCATACGAGATCTACGCAGAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SA707	 CAAGCAGAAGACGGCATACGAGATGGAGACTAAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SA708	 CAAGCAGAAGACGGCATACGAGATGTCGCTCGAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SA709	 CAAGCAGAAGACGGCATACGAGATGTCGTAGTAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SA710	 CAAGCAGAAGACGGCATACGAGATTAGCAGACAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SA711	 CAAGCAGAAGACGGCATACGAGATTCATAGACAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SA712	 CAAGCAGAAGACGGCATACGAGATTCGCTATAAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SB701	 CAAGCAGAAGACGGCATACGAGATAAGTCGAGAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SB702	 CAAGCAGAAGACGGCATACGAGATATACTTCGAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SB703	 CAAGCAGAAGACGGCATACGAGATAGCTGCTAAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SB704	 CAAGCAGAAGACGGCATACGAGATCATAGAGAAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SB705	 CAAGCAGAAGACGGCATACGAGATCGTAGATCAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SB706	 CAAGCAGAAGACGGCATACGAGATCTCGTTACAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SB707	 CAAGCAGAAGACGGCATACGAGATGCGCACGTAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SB708	 CAAGCAGAAGACGGCATACGAGATGGTACTATAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SB709	 CAAGCAGAAGACGGCATACGAGATGTATACGCAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SB710	 CAAGCAGAAGACGGCATACGAGATTACGAGCAAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SB711	 CAAGCAGAAGACGGCATACGAGATTCAGCGTTAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT	v4.SB712	 CAAGCAGAAGACGGCATACGAGATTCGCTACGAGTCAGTCAGCCGGACTACHVGGGTWTCTAAT				59	Appendix	E:	Mothur	batch	code	and	final	reads	summary	Mothur	batch	code	pcr.seqs(fasta=silva.bacteria.fasta,	start=11894,	end=25319,	keepdots=F,	processors=8)	system(rename	silva.bacteria.pcr.fasta	silva.v4.fasta)	summary.seqs(fasta=silva.v4.fasta)		make.contigs(file=stability.files,	processors=2)	summary.seqs(fasta=current)	screen.seqs(fasta=stability.trim.contigs.fasta,	group=stability.contigs.groups,	maxambig=0,	maxlength=275)	unique.seqs(fasta=stability.trim.contigs.good.fasta)	count.seqs(name=stability.trim.contigs.good.names,	group=stability.contigs.good.groups)	summary.seqs(count=stability.trim.contigs.good.count_table)	align.seqs(fasta=stability.trim.contigs.good.unique.fasta,	reference=silva.v4.fasta)	summary.seqs(fasta=current,	count=current)	screen.seqs(fasta=stability.trim.contigs.good.unique.align,	count=stability.trim.contigs.good.count_table,	summary=stability.trim.contigs.good.unique.summary,	start=1968,	end=11550,	maxhomop=8)	summary.seqs(fasta=current,	count=current)	filter.seqs(fasta=stability.trim.contigs.good.unique.good.align,	vertical=T,	trump=.)	unique.seqs(fasta=stability.trim.contigs.good.unique.good.filter.fasta,	count=stability.trim.contigs.good.good.count_table)	pre.cluster(fasta=stability.trim.contigs.good.unique.good.filter.unique.fasta,	count=stability.trim.contigs.good.unique.good.filter.count_table,	diffs=2)	chimera.uchime(fasta=stability.trim.contigs.good.unique.good.filter.unique.precluster.fasta,	count=stability.trim.contigs.good.unique.good.filter.unique.precluster.count_table,	dereplicate=t)	remove.seqs(fasta=stability.trim.contigs.good.unique.good.filter.unique.precluster.fasta,	accnos=stability.trim.contigs.good.unique.good.filter.unique.precluster.denovo.uchime.accnos)	summary.seqs(fasta=current,	count=current)	classify.seqs(fasta=stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.fasta,	count=stability.trim.contigs.good.unique.good.filter.unique.precluster.denovo.uchime.pick.count_table,	reference=trainset9_032012.pds.fasta,	taxonomy=trainset9_032012.pds.tax,	cutoff=80)	remove.lineage(fasta=stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.fasta,	count=stability.trim.contigs.good.unique.good.filter.unique.precluster.denovo.uchime.pick.count_table,		60	taxonomy=stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.pds.wang.taxonomy,	taxon=Chloroplast-Mitochondria-unknown-Archaea-Eukaryota)	dist.seqs(fasta=stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.fasta,	cutoff=0.20)	cluster(column=stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.dist,	count=stability.trim.contigs.good.unique.good.filter.unique.precluster.denovo.uchime.pick.pick.count_table)	make.shared(list=stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.list,	count=stability.trim.contigs.good.unique.good.filter.unique.precluster.denovo.uchime.pick.pick.count_table,	label=0.03)	classify.otu(list=stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.list,	count=stability.trim.contigs.good.unique.good.filter.unique.precluster.denovo.uchime.pick.pick.count_table,	taxonomy=stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.pds.wang.pick.taxonomy,	label=0.03)		system(rename	stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.shared	stability.an.shared)	system(rename	stability.trim.contigs.good.unique.good.filter.unique.precluster.pick.pick.an.unique_list.0.03.cons.taxonomy	stability.an.cons.taxonomy)	count.groups(shared=stability.an.shared)	sub.sample(shared=stability.an.shared,	size=508)	rarefaction.single(shared=stability.an.shared,	calc=sobs,	freq=100)	summary.single(shared=stability.an.shared,	calc=nseqs-coverage-sobs-invsimpson,	subsample=508)	dist.shared(shared=stability.an.shared,	calc=thetayc-jclass-braycurtis,	subsample=508)	tree.shared(phylip=stability.an.braycurtis.0.03.lt.ave.dist)			 		61	Final	reads	summary		BCCA1	 BCCA2	 BCCA3	 BCCA4	 BCCA5	 BCCA6	 BCCA7	21107	 58477	 37958	 8186	 3244	 2708	 2588		 	 	 	 	 	 	BCCA8	 BCCA9	 BCCA10	 BCCA11	 BCCA12	 BCCA13	 BCCA14	60185	 38280	 11808	 53470	 38479	 14841	 11953		 	 	 	 	 	 	BCCA15	 BCCA16	 BCCA17	 BCCA18	 BCCA19	 BCCA20	 BCCA21	42944	 77821	 44420	 5349	 18721	 20620	 31625		 	 	 	 	 	 	BCCA22	 BCCA23	 BCCA24	 BCCA25	 BCCA26	 BCCA27	 BCCA28	51064	 17915	 24291	 22451	 14089	 29582	 6795		 	 	 	 	 	 	BCCA29	 BCCA30	 BCCA31	 BCCA32	 BCCA33	 BCCA34	 BCCA35	2347	 23090	 51708	 20457	 76682	 49302	 59916		 	 	 	 	 	 	BCCA36	 BCCA37	 BCCA38	 BCCA39	 BCCA40	 BCCA41	 BCCA42	2820	 50535	 5606	 65967	 51100	 66927	 53827		 	 	 	 	 	 	BCCA43	 BCCA44	 BCCA45	 BCCA46	 BCCA47	 BCCA48	 BIDC1	45016	 70907	 36519	 46506	 65976	 77311	 7634		 	 	 	 	 	 	BIDC2	 BIDC3	 BIDC4	 BIDC6	 BIDC7	 BIDC8	 BIDC9	69152	 17863	 30845	 52387	 8192	 88031	 10751		 	 	 	 	 	 	BIDC10	 BIDC14	 BIDC15	 BIDC16	 BIDC17	 BIDC18	 BIDC19	6313	 13173	 51242	 33889	 81797	 26556	 51578		 	 	 	 	 	 	BIDC20	 BIDC21	 BIDC22	 BIDC23	 BIDC24	 BIDC25	 BIDC26	52515	 43241	 22050	 6129	 812	 65295	 43935		 	 	 	 	 	 	BIDC27	 BIDC28	 BIDC29	 BIDC30	 BIDC31	 BIDC32	 BIDC33	60270	 8606	 45720	 13747	 19667	 8025	 6105				62	Appendix	F:	Post-hoc	Tests	Table 7. Dunn’s multiple comparison test of phylum Bacteriodetes Dunn's Multiple Comparison Test Difference in rank sum Significant? P < 0.05? Summary HIV+COPD- vs HIV+COPD+ -15.21 No ns HIV+COPD- vs HIV-COPD- -24.07 Yes ** HIV+COPD- vs HIV-COPD+ -20.57 Yes * HIV+COPD+ vs HIV-COPD- -8.857 No ns HIV+COPD+ vs HIV-COPD+ -5.357 No ns HIV-COPD- vs HIV-COPD+ 3.5 No ns  Table 8. Dunn’s multiple comparison test of phylum Proteobacteria Dunn's	Multiple	Comparison	Test	 Difference	in	rank	sum	 Significant?	P	<	0.05?	 Summary	HIV+COPD-	vs	HIV+COPD+	 1.357	 No	 ns	HIV+COPD-	vs	HIV-COPD-	 22.37	 Yes	 **	HIV+COPD-	vs	HIV-COPD+	 19.74	 Yes	 *	HIV+COPD+	vs	HIV-COPD-	 21.01	 No	 ns	HIV+COPD+	vs	HIV-COPD+	 18.39	 No	 ns	HIV-COPD-	vs	HIV-COPD+	 -2.625	 No	 ns		Table 9. Dunn’s multiple comparison test of phylum Firmicutes Dunn's	Multiple	Comparison	Test	 Difference	in	rank	sum	 Significant?	P	<	0.05?	 Summary	HIV+COPD-	vs	HIV+COPD+	 0.5714	 No	 ns	HIV+COPD-	vs	HIV-COPD-	 -13.15	 No	 ns	HIV+COPD-	vs	HIV-COPD+	 -23.15	 Yes	 **	HIV+COPD+	vs	HIV-COPD-	 -13.73	 No	 ns	HIV+COPD+	vs	HIV-COPD+	 -23.73	 Yes	 *	HIV-COPD-	vs	HIV-COPD+	 -10	 No	 ns			 		63	Table 10. Dunn’s multiple comparison test of Shannon Diversity Dunn's	Multiple	Comparison	Test	 Difference	in	rank	sum	 Significant?	P	<	0.05?	 Summary	HIV+COPD-	vs	HIV+COPD+	 1.357	 No	 ns	HIV+COPD-	vs	HIV-COPD+	 -17.05	 No	 ns	HIV+COPD-	vs	HIV-COPD-	 -14.76	 No	 ns	HIV+COPD+	vs	HIV-COPD+	 -18.41	 No	 ns	HIV+COPD+	vs	HIV-COPD-	 -16.12	 No	 ns	HIV-COPD+	vs	HIV-COPD-	 2.292	 No	 ns		Table 11. Dunn’s multiple comparison test of richness Dunn's	Multiple	Comparison	Test	 Difference	in	rank	sum	 Significant?	P	<	0.05?	 Summary	HIV+COPD-	vs	HIV+COPD+	 6.857	 No	 ns	HIV+COPD-	vs	HIV-COPD+	 -18.69	 Yes	 *	HIV+COPD-	vs	HIV-COPD-	 -19.86	 Yes	 *	HIV+COPD+	vs	HIV-COPD+	 -25.55	 Yes	 *	HIV+COPD+	vs	HIV-COPD-	 -26.72	 Yes	 *	HIV-COPD+	vs	HIV-COPD-	 -1.167	 No	 ns		Table 12. Dunn’s multiple comparison test of evenness Dunn's	Multiple	Comparison	Test	 Difference	in	rank	sum	 Significant?	P	<	0.05?	 Summary	HIV+COPD-	vs	HIV+COPD+	 -11.79	 No	 ns	HIV+COPD-	vs	HIV-COPD+	 -5.435	 No	 ns	HIV+COPD-	vs	HIV-COPD-	 1.274	 No	 ns	HIV+COPD+	vs	HIV-COPD+	 6.351	 No	 ns	HIV+COPD+	vs	HIV-COPD-	 13.06	 No	 ns	HIV-COPD+	vs	HIV-COPD-	 6.708	 No	 ns				64	Appendix	G:	Lists	of	GO	Pathways	from	DMRs	Enrichment	Table 13. List of GO pathways in biological process that associated with HIV infection GO	pathway:	Biological	Process	Number	of	Genes	 p-value	pattern	specification	process	 68	 2.34E-09	nervous	system	development	 191	 3.63E-09	cell	development	 164	 2.38E-08	homophilic	cell	adhesion	 31	 2.80E-08	cellular	developmental	process	 272	 9.97E-07	system	development	 326	 2.01E-06	cell	differentiation	 255	 3.69E-06	multicellular	organismal	development	 368	 4.02E-06	cellular	process	 1001	 8.18E-06	anatomical	structure	development	 362	 8.26E-06	anatomical	structure	morphogenesis	 202	 9.41E-06	regionalization	 45	 1.01E-05	developmental	process	 403	 1.24E-05	cell-cell	adhesion	 54	 1.74E-05	neurogenesis	 121	 3.02E-05	anatomical	structure	formation	involved	in	morphogenesis	 159	 4.75E-05	neuron	development	 89	 6.38E-05	anterior/posterior	pattern	specification	 32	 6.48E-05	generation	of	neurons	 113	 7.82E-05	cell	adhesion	 102	 9.97E-05	intracellular	protein	transport	in	other	organism	involved	in	symbiotic	interaction	 4	 1.00E-04	regulation	of	transmembrane	receptor	protein	serine/threonine	kinase	signaling	pathway	 27	 1.00E-04	extracellular	transport	 4	 1.00E-04	tube	morphogenesis	 41	 1.00E-04	symbiont	intracellular	protein	transport	in	host	 4	 1.00E-04	regulation	of	viral	protein	levels	in	host	cell	 4	 1.00E-04	intracellular	transport	of	viral	proteins	in	host	cell	 4	 1.00E-04	biological	adhesion	 102	 1.00E-04	organ	morphogenesis	 87	 2.00E-04	cellular	component	morphogenesis	 104	 2.00E-04	tube	closure	 15	 2.00E-04	neuromuscular	synaptic	transmission	 7	 2.00E-04	neuron	differentiation	 104	 2.00E-04			65	Table 13. List of GO pathways in biological process that associated with HIV infection GO	pathway:	Biological	Process	Number	of	Genes	 p-value	tissue	development	 143	 2.00E-04	transcription	from	RNA	polymerase	II	promoter	 140	 2.00E-04	cell	morphogenesis	 99	 3.00E-04	single-multicellular	organism	process	 471	 3.00E-04	negative	regulation	of	response	to	stimulus	 85	 3.00E-04	regulation	of	transforming	growth	factor	beta	receptor	signaling	pathway	 17	 3.00E-04	mesenchyme	development	 25	 4.00E-04	primary	neural	tube	formation	 15	 4.00E-04	regulation	of	transcription	from	RNA	polymerase	II	promoter	 121	 4.00E-04	tube	formation	 20	 4.00E-04	neural	tube	development	 22	 4.00E-04	cell	projection	organization	 97	 4.00E-04	multicellular	organismal	process	 471	 5.00E-04	cell	fate	commitment	 31	 5.00E-04	mesenchymal	cell	development	 21	 5.00E-04	neural	tube	closure	 14	 6.00E-04	embryo	development	 93	 6.00E-04	epithelium	development	 67	 6.00E-04	neuron	projection	development	 76	 6.00E-04	nucleobase-containing	compound	metabolic	process	 438	 7.00E-04	morphogenesis	of	an	epithelium	 45	 7.00E-04	mesenchymal	cell	differentiation	 22	 7.00E-04	negative	regulation	of	protein	serine/threonine	kinase	activity	 16	 7.00E-04	tube	development	 51	 8.00E-04	cell	morphogenesis	involved	in	differentiation	 75	 9.00E-04	epithelial	tube	morphogenesis	 32	 9.00E-04	transmembrane	receptor	protein	serine/threonine	kinase	signaling	pathway	 35	 9.00E-04	sympathetic	nervous	system	development	 8	 9.00E-04	regulation	of	cell	development	 56	 0.001	regulation	of	action	potential	 21	 0.0011	peripheral	nervous	system	neuron	development	 5	 0.0011	peripheral	nervous	system	neuron	differentiation	 5	 0.0011	RNA	metabolic	process	 336	 0.0012	hematopoietic	progenitor	cell	differentiation	 8	 0.0012			66	Table	13.	List	of	GO	pathways	in	biological	process	that	associated	with	HIV	infection	GO	pathway:	Biological	Process	Number	of	Genes	 p-value	negative	regulation	of	transmembrane	receptor	protein	serine/threonine	kinase	signaling	pathway	 16	 0.0014	negative	regulation	of	cyclin-dependent	protein	kinase	activity	 6	 0.0014	negative	regulation	of	signaling	 72	 0.0015	transforming	growth	factor	beta	receptor	signaling	pathway	 24	 0.0016	purine	nucleotide	catabolic	process	 51	 0.0016	single-organism	process	 618	 0.0016		Table 14. List of GO pathways in cellular component that associated with HIV infection GO	pathway:	Cellular	Component	Number	of	Genes	 p-value	intracellular	part	 932	 5.23E-10	intracellular	 951	 9.67E-10	intracellular	organelle	 823	 6.90E-09	nucleus	 500	 1.17E-08	organelle	 823	 9.97E-09	intracellular	membrane-bounded	organelle	 751	 1.67E-08	membrane-bounded	organelle	 751	 2.16E-08	cell	 1067	 7.74E-08	cell	part	 1067	 7.51E-08	cytoplasm	 711	 3.58E-07	membrane-enclosed	lumen	 292	 8.24E-06	organelle	lumen	 287	 1.39E-05	intracellular	organelle	lumen	 282	 2.28E-05	nuclear	part	 262	 6.35E-05	microtubule	cytoskeleton	 89	 7.74E-05	nucleoplasm	 139	 7.86E-05	axon	 37	 2.00E-04	nuclear	lumen	 234	 2.00E-04	axon	part	 21	 3.00E-04	neuron	projection	 67	 6.00E-04	intracellular	organelle	part	 516	 7.00E-04	organelle	part	 522	 7.00E-04	synapse	 53	 7.00E-04	intracellular	non-membrane-bounded	organelle	 306	 9.00E-04	neuronal	cell	body	 35	 9.00E-04		67	Table 14. List of GO pathways in cellular component that associated with HIV infection non-membrane-bounded	organelle	 306	 9.00E-04	nucleoplasm	part	 80	 0.001	cytoplasmic	part	 517	 0.0011	cell	body	 36	 0.0016	microtubule	organizing	center	 51	 0.0018	proteasome	core	complex	 6	 0.0018	terminal	button	 9	 0.0023	cytoskeleton	 153	 0.0025	cytoskeletal	part	 112	 0.0056	membrane	coat	 12	 0.0056	coated	membrane	 12	 0.0056																				68		Table 15. List of GO pathways in molecular function that associated with HIV infection GO	pathway:	Molecular	Function	Numbers	of	Genes	 p-value	binding	 937	 2.65E-12	sequence-specific	DNA	binding	transcription	factor	activity	 123	 1.43E-09	nucleic	acid	binding	transcription	factor	activity	 123	 1.62E-09	sequence-specific	DNA	binding	 86	 1.01E-07	DNA	binding	 224	 2.68E-07	nucleic	acid	binding	 299	 8.61E-07	regulatory	region	DNA	binding	 48	 2.71E-06	regulatory	region	nucleic	acid	binding	 48	 2.71E-06	protein	binding	 593	 4.26E-06	transcription	regulatory	region	DNA	binding	 46	 7.16E-06	organic	cyclic	compound	binding	 448	 4.18E-05	heterocyclic	compound	binding	 444	 4.39E-05	cation	binding	 342	 6.50E-05	metal	ion	binding	 332	 1.00E-04	ion	binding	 470	 2.00E-04	calcium	ion	binding	 71	 4.00E-04	identical	protein	binding	 87	 5.00E-04	transcription	regulatory	region	sequence-specific	DNA	binding	 22	 5.00E-04	protein	complex	binding	 36	 0.001	oligopeptide	transporter	activity	 4	 0.0014	core	promoter	sequence-specific	DNA	binding	 8	 0.0014	core	promoter	binding	 10	 0.0017	threonine-type	peptidase	activity	 6	 0.0027	threonine-type	endopeptidase	activity	 6	 0.0027		

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0307415/manifest

Comment

Related Items