Pro-inflammatory cytokines in aqueous humor from ADAMTS10-mutant Beagles at an early stage of open-angle glaucoma (OAG)

Abstract

Objective.

To investigate the level of pro-inflammatory cytokines in aqueous humor from ADAMTS10-mutant research Beagles at an early stage of open-angle glaucoma (OAG).

Animals Studied.

Twenty-four research Beagles were enrolled into this case-controlled study. Three groups of ADAMTS10-mutant research Beagles were enrolled in this study: (1) Intraocular pressure (IOP) 15–25 mmHg (normotensive IOP), (2) IOP 25–30 mmHg, and (3) IOP 30–36 mmHg. An unaffected control group of Beagles that were either ADAMTS10-wildtype (WT) or heterozygote-carrier with normotensive IOPs were also enrolled.

Procedures.

Aqueous humor samples were collected and analyzed for 16 pro-inflammatory cytokines using the Canine Cytokine SpikeMix^™ and target mass spectrometry via multiple reaction monitoring (MRM-MS). Statistical differences between the four groups’ pro-inflammatory cytokines, as well as correlations between pro-inflammatory cytokines and IOPs, were analyzed using Kruskal-Wallis tests and Spearman’s rho (ρ) correlations, respectively.

Results.

IOP 15–25 mmHg (n=8); IOP 25–30 mmHg (n=6); IOP 30–36 mmHg (n=5); and Control (WT n=3, carrier n=2, total n=5). There were no differences between pro-inflammatory cytokines across the four groups (all p-values > 0.05). IL-13 had a moderate positive correlation with IOP, but was non-significant (ρ=0.373, p=0.073); IL-1β had a moderate negative correlation with IOP but was also non-significant (ρ=−0.344, p = 0.100). All other pro-inflammatory cytokines had only mild correlation with IOPs (|ρ| < 0.229, p > 0.05).

Conclusions.

There were no significant changes in the investigated pro-inflammatory cytokines with elevated IOP in canine ADAMTS10-OAG.

INTRODUCTION

Glaucomatous optic neuropathy is a common cause of blindness in humans and canines with intraocular pressure (IOP)-related biomechanical stress being a major risk factor.^1–11 Primary open-angle glaucoma (OAG) is characterized by a normal appearing iridocorneal angle and, in some cases, gradual elevation in IOP.^1,3,12 As the disease progresses, the angle subsequently collapses.¹³

Currently, all known forms of canine OAG are caused by loss of function mutations in either ADAMTS10 or ADAMTS17.^14–19 The ADAMTS10 protein is expressed in the trabecular meshwork (TM) where manifestation of OAG begins with extracellular plaque formation and reduced aqueous humor (AH) outflow.^17,20 While the primary mechanism for elevation in IOP of OAG lies within the AH outflow pathways, early diagnosis of glaucoma is important for a better prognosis and appropriate treatment to prevent vision loss.^2,3,6 In humans with primary glaucoma, vision loss is manageable and can be prevented with early diagnosis and intervention, but this differs from many canine patients where vision loss is unfortunately inevitable.

Chronic inflammation at the trabecular meshwork affects both aqueous production and resistance to outflow, and has been associated with elevated IOP in glaucoma.^5–7,10 The level of pro-inflammatory cytokines in aqueous humor from humans with end-stage OAG has shown to have a potential role in the pathogenesis and development of OAG.^5,6,9–11 Tumor necrosis factor α (TNF-α) and transforming growth factor β-two (TGF-β2) are two of these pro-inflammatory cytokines that have been found to increase with the severity of the glaucomatous neuropathy in OAG patients.^5,6,8,10,21 Other cytokines including TGF-β1, Interleukin (IL)-8, and serum amyloid A (SAA) may play a role in ECM production, remodeling, and reduction in aqueous humor outflow at the TM in eyes with OAG and exfoliation glaucoma.¹⁰ IL-12 and IL-13 have been reported as potential early biomarkers in humans with OAG, whereas IL-10 and ostepontin (OPN) may suggest involvement of an inflammatory and oxidative stress pathway to the pathogenesis of OAG.^5–7,8,11 These cytokines have been collected from human aqueous humor at end-stage of the disease where intervention was needed to lower the IOP.^8–10,21

Additionally, studies have shown that the number of topical antiglaucoma medications (i.e. prostaglandin analogs, β-blockers, carbonic anhydrase inhibitors) is correlated with intraocular inflammation in human glaucoma patients, with an increase in proinflammatory cytokines such as IL1-β, IL-6, IL-8, IL-10, IL-12, and TNF-α.^10,22,23 The use of prophylactic topical anti-inflammatory medications in these patients has therefore been proposed to be beneficial.^10,13,24,25 Evaluating cytokines in AH from patients with an early stage of OAG can be difficult since surgical procedures that often lead to AH collection is not necessary in these patients. Although primary angle-closure glaucoma (PACG) is the most common form of primary glaucoma in canines, ADAMTS10-OAG has been studied in great detail thanks to two colonies of purpose-bred affected dogs.^3,17 These animals can be used to investigate pro-inflammatory cytokines and potential biomarkers in the AH of canines with an early stage of OAG.

In a recent canine study, target mass spectrometry via multiple reaction monitoring (MRM-MS) using Canine Cytokine Spikemix^™ has been shown to be a method to detect pro-inflammatory cytokines in canine AH. IL-18 and TNFα were elevated in canine primary angle closure glaucoma (PACG) patients compared to post-operative hypertension (POH) patients.^2,26 This study among others suggests there may be an inflammatory component in eyes from dogs with PACG.^6–8,26,27

The purpose of this study was to investigate the level of pro-inflammatory cytokines in AH from ADAMTS10-mutant research Beagles at an early state of OAG. The AH was analyzed with Canine Cytokine SpikeMix^™ using MRM-MS and the enrolled dogs were not started on any topical medications for their OAG. A group of Beagles that were either wildtype or heterozygote for the ADAMTS10 gene were used as a control group.

MATERIALS AND METHODS

Study Design and Animals

This was a case-controlled study that was performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research guidelines and was approved by the Michigan State University’s Institutional Animal Care Use Committee (IACUC).

Three groups of ADAMTS10-mutant research Beagles were enrolled in this study as follows: (1) Intraocular pressure (IOP) 15–25 mmHg (normotensive IOP; n=8), (2) IOP 25–30 mmHg (n=6), and (3) IOP 30–36 mmHg (n=5). An unaffected control group of Beagles that were either ADAMTS10-wildtype (WT) or heterozygote-carrier with normotensive IOPs were also enrolled (WT n=3, carrier n=2, total n=5). ADAMTS10 gene sequence was used to confirm genotypes.¹⁷ Homozygous and carriers were heterozygous for the G661R missense mutation for ADAMTS10-OAG affected Beagles, while normal Beagles were homozygous for the wild-type allele.¹ The Beagles that were enrolled in this study were not on any topical or systemic medication.

Ophthalmic examination

All dogs were examined by a board-certified veterinary ophthalmologist (AMK), using slit-lamp biomicroscopy (SL-17; Kowa Company Ltd., Tokyo, Japan) of the adnexal and anterior segment, as well as indirect ophthalmoscopy (Keeler Vantage; Keeler Instruments, Inc., Broomall, PA, USA; 28D condensing lens; Volk; Mentor, OH, USA) of the posterior segment (retina and optic nerve head). Monthly diurnal IOPs were assessed by rebound tonometry (Tono-Vet; Icare^® Finland Oy, Espoo, Finland) as part of the routine colony management.

Aqueous Humor Collection

All dogs were sedated with dose depending on clinician assessment of patient temperament and aqueous humor was obtained from all four groups of Beagles as previously described.¹ The antiemetic ondansetron (0.5–1.0 mg/kg; Aurobindo Pharma USA, Inc.) was given orally 2–4 h before the procedure. All dogs were sedated intravenously with a combination of butorphanol tartrate (0.2–0.4 mg/kg; Bayer HealthCare LLC), midazolam (0.2–0.4 mg/kg; Avet Pharmaceuticals, Inc.), and dexmedetomidine HCl (0.5 μg/kg; Zoetis, Inc.).¹ The ocular surface was flushed with dilute povidone-iodine (0.2%; Medline Industries, Inc.) prior to the procedure. Topical analgesia was applied with proparacaine HCl 0.5% ophthalmic solution (0.1 mL; Akorn, Inc.). Aqueous humor (100–200 μL) was obtained with the use of the 30G needle of an insulin syringe (BD Insulin Syringes with BD Ultra-Fine^™ needle) that was inserted at the dorsal limbus and care was taken to avoid contact with the iris and other vascularized structures in accordance with previously described protocols.^1,26,28 The needle was carefully removed and the insertion site at the limbus and neomycin-polymyxin B-dexamethasone 0.1% ophthalmic ointment was applied topically (Bausch & Lomb).¹ The nonsteroidal anti-inflammatory carprofen was administered subcutaneously (4.4 mg/kg; Zoetis, Inc.), and dexmedetomidine HCl sedation was reversed by intramuscular administration of atipamezole HCl (0.5 mg/kg; Zoetis, Inc.).¹ Samples were flash-frozen in liquid nitrogen, kept frozen at −80C, and then shipped on dry ice.²⁶

MRM-MS

Aqueous humor samples were submitted for analysis of pro-inflammatory cytokines to Colorado State University College of Veterinary Medicine and Biomedical Sciences’ Department of Microbiology, Immunology, and Pathology for the analysis with Canine Cytokine SpikeMix^™ and MRM-MS as previously described.²⁶ The Canine Cytokine SpikeMix^™ can detect 144 cytokines, but previous studies have shown that this Canine Cytokine SpikeMix^™ can be used to evaluate for 16 pro-inflammatory cytokines: IL-1α, IL-1β, IL-2, IL-3, IL-4, IL-5, IL-6, IL-8, IL12-A, IL-12B, IL-13, IL-18, interferon gamma (INFG), granulocyte macrophage colony stimulating factor2 (CSF-2), thrombopoietin (TPO), tumor necrosis factor α (TNF-α).^26,29 Sample processing, preparation, and analysis were performed as previously described.²⁶ Descriptive data from this study was reported as the normalized total peak area (nTPA) and median (range) using data imported from the online MRM-MS Skyline program (https://skyline.ms/project/home/begin.view).

Data Analysis

Data collected included age at the time of sampling, sex, most recent mean diurnal IOP measurements, and data from the MRM-MS analysis. Non-parametric analysis with the Kruskal-Wallis test was used to compare the 16 pro-inflammatory cytokine levels across the four different groups of dogs, and Spearman’s correlation was used to assess potential correlations between IOP elevation and the cytokines. A p-value < 0.05 was seen as significant.

RESULTS

There were 22 Beagles (10 males and 12 females) included in this study (24 eyes; 14 OS, 10 OD) with a mean ± standard deviation (SD) age of 4.7 ± 1.4 years. One eye from each dog was included in the study except for two dogs where both eyes were included in separate categories. There were five eyes in IOP 12–25 (mean ± SD IOP; 19.6 ± 3.3 mmHg), eight eyes in IOP 25–30 (IOP; 28.0 ± 1.2 mmHg), five eyes in IOP 30–36 (IOP; 32.7 ± 2.1 mmHg), and six eyes in Control (IOP; 15.3 ± 3.7 mmHg).

The summary of statistics for the 16 pro-inflammatory cytokine levels (normalized Total Peak Area [nTPA]) from the aqueous humor of the research Beagles are outlined in Figure 1 and as a supplement (Table s1). There was no statistical significance of cytokine level across the four different groups or relation to IOP. IL-13 had a moderate positive correlation with IOP values, but this finding was not significant (ρ=0.373, p=0.073). IL-1β had a moderate negative correlation with IOP, but this was also not significant (ρ=−0.344, p = 0.100) (Figure 2). All other pro-inflammatory cytokines had only mild correlation with IOPs (|ρ| < 0.229, p > 0.050). Kruskal-Wallis tests of cytokine levels by group and results of Spearman’s rho (ρ) correlation for IOP and cytokines are further outlined in Table 1.

Figure 1. Interleukin pro-inflammatory cytokine levels in the aqueous humor of Beagles.

Cytokine levels (mean normalized Total Peak Area [nTPA] ± standard deviation) in the aqueous humor (AH) of ADAMTS10-mutant and wildtype or carrier research Beagles including Interleukin (IL) 1α, IL-1β, IL-3, IL-4, IL-5, IL-6, IL-8, IL-12α, IL-12β, IL-13, and IL-18. There was no significant difference (p≤0.05) between the 16 pro-inflammatory cytokines and the four groups.

Figure 2. Modified correlation with no statistical significance.

(A) Cytokine levels (mean normalized Total Peak Area [nTPA] ± standard deviation) of Interleukin (IL)-1β had a negative modified correlation with increasing intraocular pressure (IOP) (ρ=−0.344, p = 0.100). This modified correlation was not statistically significant. (B) Cytokine levels (mean normalized Total Peak Area [nTPA] ± standard deviation) of IL-13 had a positive modified correlation with increasing intraocular pressure (IOP). This modified correlation was also not statistically significant (ρ=0.373, p=0.073).

Table 1.

Spearman’s rho (ρ) correlations for IOP and the 16 pro-inflammatory cytokines. IL-13 had a moderate positive correlation with IOP, but was non-significant (ρ=0.373, p=0.073). IL-1β had a moderate negative correlation with IOP but was also not significant (ρ=−0.344, p = 0.100) (Std. Dev. = standard deviation).

Intraocular Pressure (IOP)
Group	Mean ± Std. Dev.	Range (Minimum – Maximum)
1. IOP 15–25	19.6 ± 3.3	15.3–23.5
2. IOP 25–30	28.0 ± 1.2	26.7–29.8
3. IOP >30	32.7 ± 2.1	30.3–35.7
4. Control	15.3 ± 3.7	10.2–19.1
Cytokine	ρ	p-value
IL-1α	0.116	0.590
IL-1β	−0.344	0.100
IL-2	0.138	0.519
IL-3	−0.041	0.850
IL-4	−0.090	0.675
IL-5	−0.039	0.858
IL-6	−0.090	0.677
IL-8	−0.146	0.496
IL-12α	0.104	0.629
IL-12β	−0.109	0.613
IL-13	0.373	0.073
IL-18	0.223	0.296
TNF-α	0.208	0.330
TPO	0.229	0.282
IFN-γ	−0.091	0.671
CSF-2	−0.039	0.856

DISCUSSION

In the present study, 16 pro-inflammatory cytokines were successfully identified in the aqueous humor samples from ADAMTS10-mutant and ADAMTS10-wild-type or carrier research Beagles using the Cytokine SpikeMix^™ and MRM-MS. There was no statistical significance of cytokine levels (nTPA) across the four different groups or relation to IOP.

A significant increase in biomarkers of oxidative stress and inflammatory pathways have been identified in the aqueous humor of human OAG patients.^8,9,21,22 Table 2 provides a review of previous studies that detected cytokines in AH from humans and animals with glaucoma. Increased vascular permeability through the blood–aqueous and retinal barrier breakdown may potentiate oxidative stress to the anterior segment and retina ischemia in dogs with primary glaucoma as shown by iris angiography.²⁷ As the IOP increases, more ischemia develops from mechanical pressure resulting in more intraocular inflammation.^26,27 Vidal-Villegas and colleagues showed a significant increase in IL-12, IL-13, and monocyte chemoattractant protein-1 in AH from OAG affected human patients.¹¹ In our study, IL-13 had a moderate positive correlation with IOP values, but this finding was not significant. IL-13 is an anti-inflammatory cytokine that maintains and regulates IgE production and allergic responses.³⁰ It also plays a unique role in fibrosis by regulating the extracellular matrix (ECM).³⁰ The lack of significance could be related to the small sample size number or due to relative low IOPs for the dogs enrolled in our study. It could be speculated that as IOP increases in these dogs, and there is further damage to the ECM, IL-13 production may increase due to further remodeling of the ECM. It could also be speculated that IL-13 increases because of a longer duration of mild increase IOP as was seen in the 25–35 mmHg group. Future studies could focus on IL-13 and its importance in canine OAG patients.

Table 2.

Review of literature investigating aqueous humor cytokine levels in glaucomatous eyes in human, canine, feline, and mice species.

Paper	Species	Cytokines in Aqueous Humor	Positive Correlation with IOP	Type of Disease
Pumphrey et al. 20241	Canine	Elevated (p<0.05) PACG: MMP-2 inhibition	Not evaluated	PACG, OAG
Oikawa et al. 202336	Feline	Elevated (p<0.05): TGF-β2	Yes: TGF-β2	FCG
Yin et al. 202321	Human	Elevated (p<0.05): TNF-α, TNF-β2	Yes: TGF-α	OAG
Dammak et al. 20238	Human	Elevated (p<0.05): OPN, MMP-9, TNF-α, and IL-10	Not evaluated	OAG
Terhaar et al. 202226	Canine	Elevated (p<0.05) OAG: TNF-α, IL-18 Elevated (p<0.05) POH: TNF-α, IL-4, IL-6, and IL-18	Yes: IL-18	OAG
Vidal-Villegas et al. 202211	Human	Elevated (p<0.05) OAG: IL-12, IL-13 Elevated (p<0.05) PEXG: monocyte chemoattractant protein-1	Not evaluated	OAG, PEXG
Ko et al. 202237	Mice	Elevated (p<0.05): TGF-β2	Not evaluated	OAG
Pumphrey et al. 202128	Canine	Elevated (p<0.05): MMP-1, MMP-2, MMP-3, MMP-8, MMP-9, MMP-10, TIMP-1, and TIMP-2	Not evaluated	PACG
Yun et al. 20213	Canine	Elevated (p<0.05): SPP1, PGLYRP2, TWHAE, MGAM, and VIM	Not evaluated	PACG
Park et al. 202133	Human	Elevated (p<0.05) OAG: TGF-β1, TGF-β2 Elevated (p<0.05) EXG: TGF-β1, TGF-β2, IL-8, growth factors MIP-1α, fractalkine, immune cell growth fact or Flt3 ligand	Not evaluated	OAG, EXG
Burgos-Blasco et al. 202038	Human	Elevated (p<0.05): IL-5, IL-12, IL-15, IFN-γ, and MIP-1α	Not evaluated	OAG
Khalef et al. 20179	Human	Elevated (p<0.05): IL-8, TNF-α, TNF-β1, SAA Decreased (p<0.05): IL-6	Not evaluated	OAG, PEXG
Takai et al. 201222	Human	Elevated (p<0.05): IL-8, TGF-β, and SAA Decreased (p<0.05): IL-6 in OAG	Yes: TGF-β1, IL-8, SAA	OAG, EXG

MMP-9 = matrix metalloproteinase 9

OPN = osteopontin

SAA = serum amyloid A

OAG = primary open angle glaucoma

PACG = primary closed angle glaucoma

FCG = feline congenital glaucoma

PEXG = pseudoexfoliation glaucoma; EXG = exfoliation glaucoma

POH = post-operative hypertension

SPP1 = secreted phosphoprotein 1

PGLYRP2 = peptidoglycan recognition proteins 2

TWHAE = tyrosine 3-monooxygenase

MGAM = maltase-glucoamylase

VIM = vimentin

MIP = macrophage inflammatory protein

IL-1β is a major pro-inflammatory cytokine, which had a moderate negative correlation with IOP in our study, but this negative correlation was also non-significant. IL-1α, IL-1β, and IL-6 are produced by the trabecular meshwork (TM) and are released after damage or from NK-kB pathway activation.⁵ Previous studies have found TNF-β1, IL-8, and SAA to be significantly higher, whereas IL-6 was decreased in the aqueous humor from human OAG patients when compared to exfoliation and pseudoexfoliation glaucoma patients.^9,22 The decrease in IL-6 as IOP increases is thought to be related to loss of TM cells and this could potentially also be the underlying cause for the moderate negative correlation of IL-1β with increased IOP in our study.^9,22 Alternatively, the lack of significance could indicate that the negative correlation is not a true representative of the level of IL-1β in canine AH.

In this study, ocular immune privilege or tolerance may play a role in suppressing or deviation of the immune system. As the IOP increased there was a moderate decrease in the pro-inflammatory cytokine IL-1β, and a moderate increase in the anti-inflammatory cytokine IL-13. Although not significant, the immune system is working to stay balanced. The relationship between cytokines and early elevation in intraocular pressure seen in this study could be related to this anterior chamber associated immune deviation phenomenon as opposed to an active inflammatory response given the lack of elevation of other cytokines.⁷ At a certain IOP, however, the immune system may be unable to control the inflammation. Furthermore, future studies should investigate the involvement of IL-13 and IL-1β with OAG in dogs.

Although the role of intraocular inflammation before and during periods of elevated IOP in glaucoma patients is still not well understood, neuroprotective and anti-inflammatory agents are occasionally prescribed for both the affected and for the normotensive fellow eye.¹³ Studies in humans with OAG have shown that even with control of IOP, some patients have progressive retinal ganglion cells loss and vision impairment likely due to neuronal stress and glutamate release^13,31 Neutrophilic and lymphoplasmacytic inflammation as well as pigment dispersion in canine PACG also supports the importance of inflammation in glaucomatous eyes.^13,32 In a recent survey to veterinary ophthalmologists, 38.78% responded that they prescribes a topical anti-inflammatory for the normotensive fellow eye in dogs with unilateral primary glaucoma.¹³ It was presumed that the dogs were affected with PCAG; however, this was not confirmed in the questionnaire provided to the clinicians.¹³ A clinical trial by Miller et al. used the topical demecarium bromide, which is an acetylcholinesterase inhibitor, as well as topical corticosteroid (βmethasone) in PCAG affected dogs, both applied once daily, which equally delayed the onset of glaucoma in the unaffected eye.^24,27 Based on the results from our study, there was no significant elevation in cytokines that would indicate activation of an inflammatory response that would require prophylactic treatment in early stage of OAG in dogs; however, these research Beagles were affected with OAG, and not PCAG. Additional studies are still needed to identify early biomarkers before the onset of disease.

Transforming growth factor beta (TGF-β2) has shown to be elevated in the aqueous humor of glaucomatous patients in previous studies.^{9,13,17,21,22,33–35} TGF-β plays a role in altering the quantity and quality of ECM in the aqueous humor outflow pathway, which was found to have a positive correlation with IOP in glaucomatous cats homozygous for LTBP2 mutation.^8,36 It is suggested that LTBP2 mutation lead to an increase in TGF-β2 expression resulting in microfibril abnormalities.³⁶ Defective microfibrils have been suggested to be the cause for the elevation of TGF-β in glaucomatous patients.³⁵ The ADAMTS10 mutation has also been associated with an abnormality in microfibril structure and function.^17,34 Microfibrils are macromolecular aggregates located in the extracellular matrix (ECM).^2,35 Microfibril dysfunction is likely important to the pathogenesis of OAG and abnormalities are likely responsible for the increasing aqueous humor resistance since it traverses the TM.^1,2 Identification of a G661R missense mutation in the ADAMTS10 gene in dogs with OAG agrees with the microfibril hypothesis of glaucoma, which states that defective microfibrils may be an underlying cause of glaucoma.^2,15–17,35 TGF-β2 was not evaluated in our study since the MRM-MS analysis have in a previous study shown that the Canine Cytokine SpikeMix^™ could not detected this cytokine in aqueous humor from dogs.¹⁸ In future studies, TGF-β2 should be included in the cytokine analysis, and is a limitation of the current study.

A limitation for this study was the small sample size of a purpose bred research colony. Although inflammation is not seen significantly in the early stage of IOP elevation in this study, the role of inflammation as OAG progresses remains unknown, and requires close monitoring of affected patients. The IOP measurements were also performed monthly in this study, and it is possible that pressure spikes may have been missed over the course of the study. A study by Sanchez et al found that repeated measurements had no effect on IOP, but they did note that 42.62% of IOP elevations occurred outside standard consultation hours.³⁷ An IOP curve over a 30 hour period with measurements every 3 hours was recommended, as this was safe for corneal health with a time interval that was able to detect IOP elevations (pressure spikes).³⁷ This recommendation should be taken into consideration for future studies to ensure that pressure spikes are not missed.

In conclusion, this study did not find an increase in the investigated 16 pro-inflammatory cytokines in an early stage canine ADAMTS10-OAG. This could indicate that intraocular inflammation is not significantly present in the early stage of IOP elevation in these dogs. The use of prophylactic topical anti-inflammatories seems to not be necessary for the early stage of open angle glaucoma. Research should continue in search of early biomarkers and treatment for glaucoma.

Supplementary Material

Table S1

Summary of statistics and comparisons for the 16 pro-inflammatory cytokines levels (normalized Total Peak Area, nTPA) in the AH of ADAMTS10-mutant and wildtype or carrier research Beagles using the Kruskal-Wallis test.