|dc.description.abstract||BACKGROUND: Chronic kidney disease (CKD) is a major cause of morbidity and
mortality in humans, dogs and cats. These three species share many
commonalities in pathophysiology. A wide range of diagnostic tests have been
developed and validated to identify renal insufficiency as a result of nephron
compromise and loss. However, there is increasing evidence that the
sensitivities and specificities of many widely used diagnostic CKD tests are
suboptimal. This highlights the need to develop better biomarkers of renal
disease which have a true clinical high-throughput application.
Urinary extracellular vesicles (uEVs) have become a major focus of CKD
biomarker research in recent years. In health and disease, urinary exosomes
(~20-150nm) and microvesicles (30-1000nm), collectively referred to as
urinary EVs (uEVs), are released from renal cells into urine. They are believed
to be reflective of the status of the entire renal system, carrying markers of
parent cells on their surface, and translational proteins within the uEVs.
Despite their potential promise as a novel disease biomarker, little is
definitively known about their diagnostic utility in human renal disorders and
they have not been meaningfully studied in companion animals. This is in part
due to a lack of technological resolution and standardisation across the tools
that are available for uEV analysis. Harnessing the diagnostic utility of uEVs
would allow serial measurements acting somewhat like a ‘non-invasive biopsy’
and would be an important contribution to the field of nephrology. This study
aimed to further validate the role of uEVs in humans with chronic renal disease
and characterise and investigate urinary uEVs in dogs and cats with renal
disease using advanced technology to enumerate and define surface
phenotypic markers, in conjunction with analysis of a urinary miRNA, miR-21.
Methods: Urinary EVs from healthy individuals of all three species were
characterised using real time-PCR, Western-Blots, dot-blots and transmission
electron-microscopy. A panel of renal damage markers in canine and feline
renal tissue was developed using immunohistochemistry (IHC). This also
served to assess cross reactivity of antibodies between species. NanoSight
tracking analysis was used for enumeration of uEVs across species in health
and disease, as was flow cytometry. Flow cytometry also allowed for
interrogation of uEV phenotypes in a clinically applicable high-throughput
manner. An imaging flow cytometer was then used to corroborate findings by
gaining both a fluorescent signal and a brightfield image for particles >300nm.
Exosomal micro-RNA 21 in urine and serum was analysed in healthy dogs and
cats and dogs and cats with CKD by PCR.
RESULTS: uEVs in canine and feline urine were morphologically and
phenotypically characterised. uEVs in the cat and dog express similar surface
proteins, which were detected by human antibodies. Enumeration of uEVs (20-
150nm) was performed in minimally processed healthy control urine.
Centrifugation improved repeatability of urinary uEVs measurement, and one
freeze thaw did not have any significant effect on uEV concentration or size
when compared with fresh urine. A cohort of homogenous dogs with
histopathologically assessed renal tissue, were used to further validate the
repeatability of the NanoSight. A repeat sampling study was then conducted in
both healthy dogs and cats. It was determined that the uEV concentration and
size was similar over three days in both species and a 95% reference range
was calculated. A difference in uEV concentration between healthy individuals
and patients with chronic renal disease was then elucidated in the 150-250nm
uEV size range when corrected for urinary creatinine. Investigation of a large
human CKD cohort importantly revealed that uEV number was unaffected
irrespective of the level of proteinuria. Patients with diabetic nephropathy (DN)
had significantly more uEVs in the 150-250nm, but uEV number was not a
predictor of disease in relation to eGFR when assessed over three years of
patient follow up. A new correction factor, ratio of Area Under the Curve (AUC)
<150nm/ AUC =>150nm was tested against creatinine and USG which showed
poorer agreement between samples and will require further mathematical
modelling to assess its validity.
Dedicated flow cytometry and cutting edge imaging flow cytometry were used
to study potential surface phenotypic biomarkers of uEVs using a panexosomal
marker in conjunction with fluorescently labelled antibodies including
podocalyxin, Kidney injury molecule-1 (KIM-1) and Aquaporin 2 (AQP2).
Robust methods for reliable analysis of minimally processed uEVs by flow
cytometer (FCM) were developed in healthy urine from all three species to
ensure accurate resolution, quantification and standardisation of analysis.
Following optimisation of the analysis protocols, uEVs were then assessed in
healthy individuals of all species and those with CKD. Finally miR-21 was
assessed in urine and serum of healthy animals and in those with CKD. Data
from a small cohort of dogs indicates that miR-21 may be increased in serum
of dogs suffering from CKD.
CONCLUSION: This body of work demonstrated that canine and feline uEVs
share many features of human uEVs. In addition, uEVs can be reliably and
repeatedly enumerated by NTA in humans, dogs and cats. No evidence was
found that uEVs differ between canine and feline health individuals and
patients with CKD. Furthermore, no evidence was found that uEVs profiles
were prognostically informative in humans with CKD. Finally, data was
presented which shows preliminary evidence that urinary microRNA profiles
may be diagnostically informative. This body of work provides a wide range of
novel information which will guide the development of future CKD diagnostic
assay development. Future work should focus on machine learning for
elucidating if uEVs are predictors of renal outcome and interrogating further
microRNA targets in larger patient cohorts.||en