If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address correspondence to Mark B. Lockwood, Ph.D., M.S.N., Assistant Professor of Nursing, Department of Biobehavioral Health Science, University of Illinois Chicago, College of Nursing, 845 South Damen Avenue, Room 658 (MC 802), Chicago, IL 60612-7350.
Core Laboratory Services and Genomics and Microbiome Core Facility, Rush University Medical Center, Chicago, IllinoisDepartment of Internal Medicine, Rush University Medical Center, Chicago, Illinois
Pain, a common debilitating symptom among kidney transplant recipients (KTRs), is among the most common and undertreated symptoms after kidney transplantation.
Aims
Characterize associations between gut microbiome features and pain interference before and after kidney transplantation.
Design
Longitudinal, repeated measures study, collecting fecal specimens and pain interference data pretransplant and 3 months posttransplant.
Setting
Participants were recruited at the kidney transplant clinic at the University of Illinois Hospital & Health Sciences System.
Participants/subjects
19 living donor kidney transplant recipients.
Methods
We assessed fecal microbial community structure with shotgun metagenomic sequencing; we used pain interference scores derived from the Patient-Reported Outcomes Measurement Information System-57.
Results
We measured a reduction in the Shannon diversity index in both groups after transplantation but observed no significant differences between groups at either time point. We did observe significant differences in fecal microbial Bray-Curtis similarity index among those reporting pain interference pre- transplant versus no pain interference at 3-months posttransplant (R = .306, p = .022), and between pain interference groups at posttransplant (R = .249, p = .041). Pairwise models showed significant differences between groups posttransplant in relative abundances of several taxa, including a 5-fold reduction.ßin Akkermansia among those with pain interference and a higher relative abundance of taxa associated with chronic inflammation in those with pain interference posttransplant. Functional gene analysis identified two features that were significantly enriched in those with pain interference, including a peptide transport system gene.
Conclusions
Gut microbiota community structure differs between groups with and without pain interference at 3 months after kidney transplantation. Several taxa involved in intestinal barrier integrity and chronic inflammation were associated with posttransplant pain.
Pain, a common debilitating symptom among kidney transplant recipients (KTRs), is among the most common and undertreated symptoms after kidney transplantation (
). The sources are multifactorial and often associated with the transplant surgery, dysregulation of key biologic pathways, bone mineral disorders secondary to decreased kidney function, and comorbidities such as obesity and diabetes (
). There is a notable gap in our understanding of the effects and trajectory of pain after kidney transplantation, clinical and environmental exacerbating factors, and potential underlying mechanisms.
Pain interference (PI) measures the consequences of pain on aspects of a person's life, including the extent to which pain prevents participation in social, cognitive, emotional, physical, and recreational activities (
). Chronic pain after kidney transplantation has been shown to worsen health-related quality of life (HRQoL) and medication/treatment adherence, reduce productivity at work and school, and increase risk of rejection, graft loss, and mortality (
observed that a slight majority of KTRs experienced a decrease in symptom burden over time; however, symptom scores plateaued at 3 months posttransplant. The greatest improvements were seen among younger, non-Latinx White men (
assessed symptom burden over 5 years after kidney transplantation and found improvement in some symptoms often associated with uremia; improvement in other symptoms remained ambiguous (
). These data support the need for more longitudinal assessments in racially and ethnically heterogeneous populations, using repeated measurements of pain and associated symptoms, to better understand the dynamic nature of pain and potential underlying mechanisms.
Microbiomics, the Brain–Gut Microbiome Axis, and Chronic Pain
The brain–gut microbiome axis, a bidirectional communication pathway between the gut and the brain via a diverse set of pathways, such as the central and enteric nervous systems, is a potential mechanism to explain the relationship between kidney disease and chronic pain (
). Symbiotic microbiota in the human gastrointestinal tract provide critical immunologic, metabolic, and endocrine regulatory functions for homeostasis of the human body (
). Microorganisms in the gut synthesize neurotransmitters and short-chain fatty acids, as well as neuroactive cytokines and chemokines. These chemicals mediate central nervous system homeostasis via the vagal pathway or by crossing the blood–brain barrier directly into the brain (
). Microbiome features may differ depending if pain is visceral, inflammatory, or neuropathic. Studies observed decreased relative abundance of the genera Bifidobacterium and Lactobacillus, and increased Firmicutes:Bacteroidetes ratios at the phylum level in visceral pain, however, the role of the microbiome in other pain types is emerging (
). Moreover, because most pain studies utilized 16s ribosomal RNA sequencing, little is known about the microbe-associated functional genes that may contribute to pain. Shotgun metagenomic sequencing provides information at the taxonomic level (who is there) and the functional gene level (what they are doing); thus, our study provides additional information on the contribution of functional changes in the gut microbiome on PI.
The dynamic temporal relationships between kidney function, chronic pain, and gut microbiome features in people with end-stage kidney disease who receive a kidney transplant are unknown. Accurately phenotyping pain in end-stage kidney disease, before and after transplantation, and investigating underlying mechanisms is essential to developing novel multimodal patient-centered pain interventions. Such interventions may include modifying gut microbiome features through pre-/probiotics, diet, mindfulness, and physical activity.
Study Objectives
The primary objective of this pilot study was to characterize microbial community structure before and after kidney transplantation and determine relationships between microbial diversity and features (i.e., microbial taxa, functional genes, and pathways) and PI. A secondary objective was to characterize differences in demographic factors, renal function (serum creatinine, serum blood, urea nitrogen, and estimated glomerular filtration rate), and HRQoL between PI groups (with PI vs. without PI).
Methods
Participant Recruitment
We enrolled a heterogeneous sample of adults scheduled to receive a living-donor kidney transplant (LDKT; n = 19). Eligibility criteria included: (1) being 18 years or older; (2) receiving a kidney transplant from a live donor; (3) speaking English; and (4) being willing to sign informed consent. Exclusion criteria were: (1) receiving a multiorgan transplant; (2) a previous transplant; (3) history of autoimmune disease requiring high-dose corticosteroids (e.g., >30 mg/day); (4) history of chronic gastrointestinal disease (e.g., irritable bowel syndrome); (5) history of antibiotic use or infection in the previous 3 months; or (6) being unwilling/unable to collect fecal specimens. Participants were recruited at the Kidney Transplant Clinic at the University of Illinois Hospital & Health Sciences System. The study was conducted in accordance with the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use's Good Clinical Practice international ethical and scientific quality standard for designing, conducting, recording, and reporting trials that involve the participation of human subjects. Immunosuppression regimens differ based on individuals’ immunologic risk. Those at high risk receive more potent induction immunosuppression at time of surgery, then receive similar maintenance (long-term) immunosuppression as those at low risk. For induction immunosuppression, study participants at low immunologic risk received basiliximab (20 mg on Day 0 and Day 4); participants at high immunologic risk received thymoglobulin (dose varied) or alemtuzumab (30 mg once).
The study protocol was approved by the Office for the Protection of Research Subjects at the University of Illinois Chicago (IRB 2017-0534).
Study Design
Using a longitudinal, observational, repeated measures design, we collected patient-reported outcomes surveys (measuring PI and HRQoL) and fecal specimens for microbiome feature analysis (Shannon diversity index, Bray-Curtis similarity index, and differential abundances of microbial taxa and functional genes) from 19 LDKT recipients at two time points: within 4 weeks pretransplant and at 3 months posttransplant. Participants completed the surveys on the same days they collected fecal specimens; they were instructed to answer questionnaires based on how they were feeling at the time. To reduce the incidence of missing pretransplant samples, we included only patients who were scheduled for LDKT, as these surgeries are scheduled in advance.
Measures
Demographic variables included age, sex assigned at birth, race and ethnicity, and education. Clinical variables included origin of kidney disease, dialysis status, type of dialysis, body mass index, and measures of kidney function (serum creatinine, serum blood, urea, nitrogen, and estimated glomerular filtration rate). All demographic and clinical variables were obtained from the electronic medical record.
Pain interference
PI was measured at two time points (pretransplant and 3 months posttransplant) using the Patient-Reported Outcomes Measurement Information System-57 (PROMIS-57, Version 2.1) (
). PROMIS-57 contains a fixed number of items from seven domains (Pain Intensity, Depression, Anxiety, Physical Function, Fatigue, Sleep Disturbance, and Ability to Participate in Social Roles and Activities), with eight questions per domain plus a 0 to10 numeric rating scale for pain intensity. PROMIS-57 (Version 2.1) has been validated for use in KTRs (
) and classified PI into two categories: “pain interference” (pain T-scores ≥55) and “no pain interference” (pain T-scores <55)—based on guidelines for interpreting PROMIS scores on the HealthMeasures website (
HRQoL was measured using the disease-specific Kidney Disease Quality of Life Short Form (KDQoL-SF; Version 1.3). The core of the KDQoL-SF is the generic 36-Item Short Form Health Survey used in the Medical Outcomes Study (
), which is a psychometrically sound generic HRQoL instrument. The KDQoL-SF includes eight subscales derived from the generic core, (e.g., General Health, Vitality, Social Functioning, Role-Emotional, and Mental Health) that result in two distinct composite scores: a Physical Composite Score and Mental Composite Score (
), resulting in subscales evaluating health-related concerns of people with kidney disease (e.g., Symptoms/Problems, Burden of Kidney Disease, Effects of Kidney Disease). Scores are standardized to a range of 0 to 100, with 0 indicating the worst possible HRQoL and 100 indicating the best.
Fecal Sample Collection, Preparation, and Sequencing
Fecal Sampling
After consenting to participate, each participant was trained to use the home-based fecal specimen collection kit, which contained one 9- by 12-inch clasped return envelope, one fecal swab collection and preservation tube with a sterile fecal specimen collection swab (Norgen Biotek Corp., ON, Canada), two feces catchers (Zymo Research, Tustin, CA, United States), one pair of latex-free gloves, one biohazard bag, and a sample request form. The Norgen Biotek fecal sample collection system allows fecal specimens to be collected at home and maintained at room temperature for up to 2 years. Written instructions with images were provided for the fecal collection procedure. Participants were contacted the week before each scheduled study visit to remind them of the upcoming appointment and review the sample collection procedure. Participants collected specimens at home prior to the upcoming pretransplant or posttransplant study visit. Upon arrival at the clinic, specimens were immediately stored at −80°C until DNA extraction and sequencing were performed.
DNA Extraction, Preparation, and Sequencing
Genomic DNA was extracted from fecal specimens using a Maxwell RSC instrument (Promega Corporation, Madison, WI, United States) for automated nucleic acid recovery. Libraries were prepared using a Nextera library preparation kit (Illumina, San Diego, CA, United States) according to manufacturer instructions. Five control specimens were used. Libraries were pooled and sequenced on an Illumina MiniSeq sequencer; based on the relative abundance of reads generated, specimens were repooled and sequenced on a high-output 300-cycle kit implemented on an Illumina NextSeq500 sequencer. Extraction, library preparation, and sequencing were performed at the Genome Research Core of the Research Resources Center at the University of Illinois Chicago.
Taxonomic Profiling
Raw reads were mapped to the National Center for Biotechnology Information nucleotide database using Centrifuge (
). Taxonomic annotations for each read were obtained using a least common ancestor algorithm and then summarized across all reads to create counts per taxon. Raw counts were normalized to percentages for relative abundance.
Functional Profiling
Raw reads were mapped to the Swiss-Prot protein database using DIAMOND (
). Gene ortholog annotations were then assigned using the consensus of aligned references and then summarized across all reads to create counts per ortholog for each sample. Higher-level summaries of ortholog functions were created using the Kyoto Encyclopedia of Genes and Genomes (KEGG) BRITE hierarchical annotation (
). Raw counts were normalized to percentages for relative abundance.
Statistical Analyses
Descriptive statistics—including mean (standard deviation [SD]), median (range), and frequency (percentage)—were performed for demographic, clinical, and patient-reported outcome variables.
Differential analyses of microbial taxa
Differential analyses of taxa as compared with experimental covariates were performed using the software package edgeR (Version 3.28.1) on raw sequence counts (
). Prior to analysis, the data were filtered to remove (a) any sequence counts annotated as chloroplast or mitochondria in origin, and (b) taxa that had less than 1,000 total sequence counts, summed across all specimens, or were present in less than 30% of specimens. Data were normalized as counts per million. Trimmed mean of M values (TMM) normalized data were then fit using a negative binomial generalized linear model using experimental covariates, and statistical tests were performed using a likelihood ratio test (i.e., glmFit and glmLRT functions in edgeR). Post hoc pairwise tests were performed using the exactTest function in edgeR. Adjusted p values (q values) were calculated using the Benjamini-Hochberg false discovery rate (FDR) correction (
Oksanen, J., Blanchet, F. G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., Minchin, P. R., O'Hara, R. B., Simpson, G. L., Solymos, P., Stevens, M. H. H., Szoecs, E., & Wagner, H. (2018). vegan: Community Ecology Package. 2.4.0 [Computer software].
). Prior to analysis, the data were rarefied to a depth of 100,000 counts per sample. The resulting Shannon indices were then modeled with the sample covariates using a generalized linear model assuming a Gaussian distribution. Significance of the model (analysis of variance, or ANOVA) was tested using the F test. Post hoc pairwise tests were performed using the Mann-Whitney test. Plots were generated in R using the ggplot2 library (
Oksanen, J., Blanchet, F. G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., Minchin, P. R., O'Hara, R. B., Simpson, G. L., Solymos, P., Stevens, M. H. H., Szoecs, E., & Wagner, H. (2018). vegan: Community Ecology Package. 2.4.0 [Computer software].
). Prior to analysis, the normalized data were square root transformed. The resulting dissimilarity indices were modeled and tested for significance with the sample covariates using the permutational multivariate analysis of variance (PERMANOVA) test (also known as ADONIS). Additional comparisons of the individual covariates were also performed using analysis of similarities (ANOSIM). Plots were generated in R using the ggplot2 library (
Differential analyses of functional gene orthologs
Differential analyses of functional gene orthologs (i.e., microbial community genes with putative identical functions) as compared with experimental covariates were performed using edgeR (Version 3.28.1) on raw sequence counts (
). Prior to analysis, the data were filtered to remove any functional gene ortholog that had less than 100 total sequence counts summed across all specimens or was present in less than 30% of specimens. For functional gene summaries, summarized data were filtered to remove any functional gene group (i.e., pathway, module, or BRITE category) that had less than 1,000 total sequence counts or was present in less than 30% of specimens. As with the differential analysis of bacterial taxa, functional gene orthologs or groups were normalized to counts per million using TMM normalization. The normalized data were fit using a negative binomial generalized linear model using the experimental covariates and tested using a likelihood ratio test. Post hoc pairwise tests were performed using the exactTest function in edgeR. Adjusted p values (q values) were calculated using the Benjamini-Hochberg FDR correction (
). Significant functional gene orthologs or groups were determined based on an FDR threshold of 5% (0.05).
Results
Sample Characteristics
The sample was racially and ethnically diverse: 47% Black, 37% White, 11% Latino/Hispanic, and 5% Asian by self-report. Most participants were men and on hemodialysis, and equal proportions received the high-risk immunosuppression protocol (Table 1).
Table 1Sample Characteristics by Baseline Pain Interference Groups
We noted a reduction in the Shannon diversity index in all fecal specimens over time regardless of PI status (Fig. 1). Using PERMANOVA, which quantifies multivariate community-level differences between groups, we saw statistically significant differences in the Bray-Curtis (dis)similarity indices of individuals with versus without PI. Within this analysis, we observed no significant between-group differences in the Bray-Curtis index of fecal specimens at baseline (pairwise ANOSIM; Fig. 2). Significant differences in the Bray-Curtis index were observed among those who reported PI at pretransplant versus at 3 months posttransplant, indicating a shift in community diversity among those with PI (R = .306, p = .022), but not among those who reported no PI at pretransplant versus at 3 months posttransplant (R = −.072, p = .764). Moreover, Bray-Curtis indices were significantly different between PI groups at 3 months posttransplant (R = .249, p = .041).
Figure 1Shannon diversity index in pain interference group by visit (genus level). We observed reductions in Shannon indices in both groups, though differences were not statistically significant at either timepoint. The median Shannon index and interquartile range, a) for those with no pain interference at baseline vs. 3 month was 2.82 (IQR 2.52-3.67) vs. 2.39 (IQR 2.09-2.80) respectively, and b) for those with pain interference was 3.33 (IQR 3.09-3.58) vs. 2.50 (IQR 2.28-3.82), respectively.
Figure 2Multidimensional scaling plot of fecal microbial community in study participants at (a) baseline and (b) 3 months posttransplant. Samples are color coded by Patient-Reported Outcomes Measurement Information System-57 (PROMIS-57) pain scores. R and p represent results of analysis of similarities (ANOSIM) analysis. Shotgun metagenomic analysis of fecal microbial community data are reported at the genus level.
Taken together, these findings suggest that the community structure of the gut microbiome of participants with no PI was more stable. Pairwise models adjusted for FDR (p < .05) showed significant differences in the relative abundances of several taxa between those with PI versus no PI at 3 months posttransplant (Fig. 3). Notably, there was a 5-fold reduction of bacteria from the genus Akkermansia among participants with PI, versus those without, at posttransplant. Several taxa associated with conditions involving chronic inflammation (e.g., Mobiluncus, Enterococcus, Actinotigum) were enriched among the PI group (
Figure 3False discovery rate–corrected pairwise comparison of differential abundances of taxonomic composition in feces among adults reporting pain interference (PI) versus no PI at 3 months posttransplant. Statistically significant differences were found in the relative abundance of several microbial taxa between PI groups. There was a 5-fold reduction of microbiota from the genus Akkermansia among those with PI. Higher abundance of several microbial taxa associated with conditions that involve chronic inflammation (e.g., Mobiluncus, Enterococcus, Actinotigum) was found in those with PI. Negative values indicate taxa with lower relative abundances; positive values indicate taxa with significantly higher relative abundance.
Two functional genes (K19229, sapD-cationic peptide transport system ATP-binding protein; K07076, uncharacterized protein) appeared in greater abundance (greater than 5-fold and greater than 3-fold, respectively) in those reporting baseline PI versus no PI. No between-group differences were noted at posttransplant or in modules or pathways in pairwise comparisons.
Pain Interference, Kidney Function, and HRQoL
Several between-group differences in mean scores of kidney function were observed, though none reached statistical significance. Notably, mean estimated glomerular filtration rates were slightly lower in participants reporting PI (vs. no PI) at baseline and posttransplant (8.65 vs. 11.54 ml/min/1.73 m2 and 45.76 vs. 58.40 ml/min/1.73 m2, respectively). Mean serum creatinine was similar in both groups at baseline and posttransplant, and mean serum blood urea nitrogen levels normalized in both groups by 3 months posttransplant (Table 2).
Table 2Pairwise Comparisons of Kidney Function and Health-Related Quality of Life by Pain Interference (n = 19)
Lower scores on PROMIS-57 indicate lower severity of the selected domain, with the exception of physical health and satisfaction with participation in social roles, where lower scores indicate higher severity. M = mean; SD = standard deviation; KDQoL-SF = Kidney Disease Quality of Life Short Form; PROMIS = Patient-Reported Outcomes Measurement Information System.
c Higher scores on KDQoL-SF indicate better quality of life.
d Lower scores on PROMIS-57 indicate lower severity of the selected domain, with the exception of physical health and satisfaction with participation in social roles, where lower scores indicate higher severity.M = mean; SD = standard deviation; KDQoL-SF = Kidney Disease Quality of Life Short Form; PROMIS = Patient-Reported Outcomes Measurement Information System.
At baseline, 58% (n = 11) of participants reported PI. At posttransplant, 21% reported improvements in PI, whereas 37% (n = 7) continued to report PI. Differences in HRQoL were observed between those with and without PI. Those reporting PI at baseline were significantly more likely to report lower scores on the subdomain Burden of Kidney Disease prior to transplant, reflecting lower quality of life (19.89 vs. 62.50, p = .001). Improvements in KDQoL-SF subscales were seen at 3 months posttransplant in both groups. Statistically significant differences were observed between groups in several KDQoL-SF subdomains, including Symptom/Problems, Effects of Kidney Disease, Burden of Kidney Disease, and Energy/Fatigue. Improvements in the Physical Composite Score were seen among those with no PI. Table 2 summarizes the kidney function and HRQoL measures.
Several PROMIS-57 domains, including Fatigue and Depression, had statistically significant differences in mean scores pre- and posttransplant.
Discussion
This study leveraged the kidney transplant model to better understand the relationship between kidney function, gut microbiome features, and PI in a population with end-stage kidney disease before and after a LDKT. A key finding was differences in multiple microbiome features, specifically the Bray-Curtis (dis)similarity index, and relative abundance of several taxa at the genus level between people who reported PI versus those who did not. These features/microbiota taxa may serve as useful biomarkers or targets for patient-centered interventions.
Pain Interference and Gut Microbiome
Most participants reported PI at baseline, and a nontrivial number of participants reported PI remaining at 3 months posttransplant. Consistent with previous reports (
), pain had a significant negative effect on participants’ ability to participate in everyday physical and social activities, and a negative effect on HRQoL. Deficits related to pain have serious consequences at the individual and societal level, as evidenced by the disparate scores between PI groups on the Effects of Kidney Disease subscale of the KDQoL-SF, which captures how much kidney disease interferes with daily life, takes up time, causes frustration, or makes the respondent feel like a burden. Similar differences were seen between PI groups on the PROMIS-57 subscales for Physical Functioning and Satisfaction With Participation in Social Role and Responsibilities. It should be noted that recipients of living-donor transplants often spend less time on dialysis pre-transplant and have lower symptom burden (
); thus, the effects of PI may have been attenuated in this study. Future studies should include recipients of kidneys from deceased donors.
Research identifying associations between gut microbiota features and posttransplant outcomes is underway. Most of this research centers on transplant-specific outcomes, such as graft loss, immunosuppression dosing, and posttransplant infections (
Impact of maintenance immunosuppressive therapy on the fecal microbiome of renal transplant recipients: Comparison between an everolimus- and a standard tacrolimus-based regimen.
). One study of 61 KTRs characterized differences in microbial structure of pre- and posttransplant fecal specimens and found that a reduction in diversity of the gut microbiota was associated with organ rejection and posttransplant infections (
). Another study, with 26 participants, found distinct differences in structure and composition of the gut microbiota communities between KTRs who experienced acute rejection, urinary tract infections, and posttransplant diarrhea and those who did not (
). For example, at the genus level, Bacteroides, Ruminococcus, Coprococcus, and Dorea were significantly lower among individuals with posttransplant diarrhea versus those without. Moreover, multiple studies have shown gut microbiota are involved in immunosuppressant drug metabolism (
Impact of maintenance immunosuppressive therapy on the fecal microbiome of renal transplant recipients: Comparison between an everolimus- and a standard tacrolimus-based regimen.
). Despite these important advancements in our understanding of the associations of the microbiome with patient outcomes after kidney transplantation, the relationships between kidney function, gut microbiome features, and PI before and after LDKT were previously unknown.
Generally, reductions in microbiota diversity have been associated with chronic diseases, including those involving pain (
). Among this study's participants, a reduction in the Shannon diversity index was noted in all fecal specimens over time regardless of PI status, likely due to prophylactic antibiotics/antivirals, potent immunosuppression, and trauma from the surgical procedure itself. These findings are consistent with previous research (
). Differences in Bray-Curtis similarity indices, which demonstrate dissimilarity between specimens, were observed in pairwise comparisons between participants with no PI at baseline and those with PI at 3 months posttransplant. We also observed significant differences in beta diversity between PI groups at posttransplant. Taken together, these findings demonstrate that the community structure of gut microbiota differs between KTRs with PI versus those with no PI; this may indicate less resilience of the gut microbiome after transplantation among those experiencing PI.
A key finding of this pilot study was the relationship between the genus Akkermansia and PI before and after kidney transplantation: We noted a 5-fold reduction in Akkermansia in the PI group versus those without. Akkermansia is well-known for its probiotic effects and has been implicated in other pain conditions (
An open-labeled study on fecal microbiota transfer in irritable bowel syndrome patients reveals improvement in abdominal pain associated with the relative abundance of akkermansia muciniphila.
). It plays an important role in mediating systemic inflammation by maintaining the integrity of the intestinal barrier and preventing translocation of neurotoxic metabolites (
). Because of its probiotic properties, Akkermansia may serve as a target for intervention.
We identified other taxa at the genus level that were at increased relative abundances in the PI group and have been associated with inflammation. Murdochiella are gram-positive bacteria that were at greater than 6-fold higher abundance in the PI group. These bacteria have been isolated from patient wounds and produce significant amounts of lactic acid and moderate amounts of acetic, butyric, and succinic acid (
Tryptophan metabolism, its relation to inflammation and stress markers and association with psychological and cognitive functioning: Tasmanian Chronic Kidney Disease pilot study.
). Interestingly, while the genus Mobiluncus is often associated with bacterial vaginosis, its role is not clearly understood. Recent case studies of Mobiluncus curtisii bacteremia in men have suggested a possible intestinal origin (
). Further investigation of these genera as potential mediators of pain in chronic kidney disease is warranted.
Kidney Function and Gut Microbiome
We were unable to detect any statistically significant differences in estimated glomerular filtration rate, serum creatinine, or blood urea nitrogen between PI groups at baseline or 3 months posttransplant. This is consistent with previous studies, and suggests that surrogate markers of kidney function may not correlate with gut microbiome features (
). However, given the small sample size and numerous confounding factors in this population, these findings should be confirmed.
Functional Genes
Only two functional genes were differentially abundant by PI group: K19229 and K07076. K19229 is a cationic peptide transport system ATP-binding protein that is part of a complex that includes sapA (K19226), sapB (K19227), sapC (K19228), and sapF (K19230). Researchers have shown that expression and trafficking of ATP-binding cassette transporters can be modified by pain and/or opioid pharmacotherapy (
); more research is needed to understand the role of K19229 in kidney disease. K07076 is annotated as an uncharacterized protein; this makes it hard to determine its role in the microbiome and relationship to PI. The relatively few functional gene differences between groups may reflect functional redundancy in microorganisms in the mammalian gastrointestinal tract (e.g.,
), leading to greater numbers of differentially abundant features observed in taxonomic analysis.
Limitations
First, the small sample size may not be sufficiently powered to detect statistically significant differences in some variables. However, we did see statistically significant effects between PI scores and several gut microbiome features. Second, factors involved in chronic kidney disease and kidney transplantation are complex, and we could not control for all potential confounders. However, during the screening process, we controlled for a number of important covariates (e.g., recent antibiotic use) known to affect the microbiome. Finally, we collected no data on diet, an important mediator of the microbiome. Future studies should be designed to simultaneously control for multiple key factors (e.g., diet, age, concomitant medications) that mediate microbiome composition and function.
Conclusion
This study provides data on changes in the gut microbiome as a mechanism for pain before and after kidney transplantation. The results support recent evidence that symptom burden, specifically burden related to PI, remains for a nontrivial number of KTRs and extends previous findings to include the relationship between microbiome features and PI after restoration of kidney function via a LDKT. These findings support the need for a sufficiently powered study in a larger KTR population with high symptom burden (e.g., deceased-donor recipients). Rigorous symptom phenotyping after kidney transplantation is needed. Once the relationship between gut microbiome features and PI is further confirmed, patient-centered interventions to optimize the microbiome can be developed and tested.
Clinical Implications
These findings may provide motivation for KTRs experiencing pain to participate in lifestyle modification-based interventions (e.g., diet modification, mindfulness, exercise) to promote optimal gut microbiome health to reduce pain. Nurses will play a critical role in developing patient-centered interventions and educational resources to aid KTRs in optimizing their gut microbiome health.
Conflict of interest
None.
Acknowledgments
The authors thank Anita Pakrasi, Lor Lawson, Karen Lostaunau, Shree Patel, Jamie Belken, Patricia West-Thielkie, and the UIH transplant team and all of the study participants for making this study possible. We also thank Kyra Freestar and Rica Asuncion-Reed of Bridge Creek Editing for providing editorial support for various drafts of this manuscript. Additional editing was provided by Kevin Grandfield, Publications Manager for the University of Illinois Chicago Department of Biobehvioral Nursing Science. DNA extraction and shotgun metagenomic sequencing were performed at the Genomics Core at the University of Illinois; basic processing of the raw data was performed by the University of Illinois Chicago Research Informatics Core (UICRIC). ML & AD were supported by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health (U01DK123787). AD was supported by the National Center for Complementary and Integrative Health under Award Number K24AT011995. ML was supported by the National Institute of Nursing Research of the National Institutes of Health under Award Number K23NR018482 and L30NR020114 and the University of Illinois Chicago College of Nursing’ Dean's Fund. . The content is solely the responsibility of the authors.
References
Ahmad S.
Bromberg J.S.
Current status of the microbiome in renal transplantation.
An open-labeled study on fecal microbiota transfer in irritable bowel syndrome patients reveals improvement in abdominal pain associated with the relative abundance of akkermansia muciniphila.
Tryptophan metabolism, its relation to inflammation and stress markers and association with psychological and cognitive functioning: Tasmanian Chronic Kidney Disease pilot study.
Oksanen, J., Blanchet, F. G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., Minchin, P. R., O'Hara, R. B., Simpson, G. L., Solymos, P., Stevens, M. H. H., Szoecs, E., & Wagner, H. (2018). vegan: Community Ecology Package. 2.4.0 [Computer software].
Impact of maintenance immunosuppressive therapy on the fecal microbiome of renal transplant recipients: Comparison between an everolimus- and a standard tacrolimus-based regimen.
00177-1&id=gr1.jpg){kind=link}
00177-1&id=gr2.jpg){kind=link}
00177-1&id=gr3.jpg){kind=link}