Prospective evaluation of the lymph node proteome in dogs with multicentric lymphoma supplemented with sulforaphane

Abstract Background Lymphoma (LSA) is a common malignancy in dogs. Epigenetic changes are linked to LSA pathogenesis and poor prognosis in humans, and LSA pathogenesis in dogs. Sulforaphane (SFN), an epigenetic‐targeting compound, has recently gained interest in relation to cancer prevention and therapy. Objective Examine the impact of oral supplementation with SFN on the lymph node proteome of dogs with multicentric LSA. Animals Seven client‐owned dogs with multicentric LSA. Methods Prospective, nonrandomized, noncontrolled study in treatment‐naïve dogs with intermediate or large cell multicentric LSA. Lymph node cell aspirates were obtained before and after 7 days of oral supplementation with SFN, and analyzed via label‐free mass spectrometry, immunoblots, and Gene Set Enrichment Analysis. Results There was no clinical response and no adverse events attributed to SFN. For individual dogs, the expression of up to 650 proteins changed by at least 2‐fold (range, 2‐100) after supplementation with SFN. When all dogs where analyzed together, 14 proteins were significantly downregulated, and 10 proteins were significantly upregulated after supplementation with SFN (P < .05). Proteins and gene sets impacted by SFN were commonly involved in immunity, response to oxidative stress, gene transcription, apoptosis, protein transport, maturation and ubiquitination. Conclusions and Clinical Importance Sulforaphane is associated with major changes in the proteome of neoplastic lymphocytes in dogs.


| INTRODUCTION
The prognosis for dogs with multicentric lymphoma (LSA) has not substantially improved in the last 2 decades. The mainstay of treatment is multiagent chemotherapy, but most dogs eventually die from the disease within 6 to 12 months after diagnosis. [1][2][3][4] With the addition of radiation therapy or immunotherapy, no major improvement in the prognosis is noted when compared to chemotherapy alone. [5][6][7][8] Therefore, new treatments avenues for dogs with multicentric LSA are warranted.
Sulforaphane (SFN) is an isothiocyanate derived from cruciferous vegetables such as broccoli or cauliflower. 9,10 Sulforaphane has chemopreventive activity in multiple cancer types. 9 One of the main mechanism of action of SFN is to promote the dissociation of nuclear factor E2-related factor 2 (Nrf2) from the cytoplasmic protein Keap1, with subsequent nuclear translocation of Nrf2 and modulation of many antioxidant response element-driven genes. [9][10][11] In this manner, SFN inhibits phase 1 enzymes (ie, cytochrome P450), which convert procarcinogens into carcinogens. It also induces several phase 2 enzymes (ie, glutathione transferase), which detoxify carcinogens and facilitates their excretion from the body. 9 Epigenetic alterations refer to changes in gene expression and heritable traits that do not involve modifications of the DNA sequence. 12 For example, removal of acetyl groups from the histone tails by histone deacetylases (HDAC) results in a closed chromatin conformation and repression of gene transcription. 13 The addition of methyl groups in the promoter regions of genes by DNA methyltransferases (DNMT) interferes with the binding of transcription factors and leads to gene silencing. 12 Sulforaphane, in addition to its chemopreventive properties, also caries cancer-suppressive properties by acting as a HDAC inhibitor (HDACi) and DNMT inhibitor (DNMTi) in various cancer cells, leading to re-expression of various tumor suppressor genes. 9,[14][15][16][17][18][19][20][21] Secondary to its effects on the epigenome or by direct SFN-protein interactions, SFN also induces major changes in the proteomic profile of several cancer cells in rodents and people. [22][23][24] Since epigenetic events are reversible, various agents targeting the epigenome have been developed for people with LSA. [25][26][27] On the other hand, although there is evidence for the presence of epigenetic dysregulations in dogs with LSA, [28][29][30][31][32][33] little research has been performed regarding the use of epigenetic-targeted treatments. A bioavailability profile of SFN in healthy dogs is comparable to humans, with a significant decrease in HDAC activity observed 24 hours after consumption of 1 dose of SFN. 34 Sulforaphane also reduces cell invasion and decreases focal adhesion kinase phosphorylation in canine osteosarcoma cell lines. 35 However, to our knowledge, the impact of SFN in cancer-bearing dogs has not been explored.
The primary objective of this study was to analyze changes in the lymph node proteome of dogs with treatment-naïve multicentric LSA before and after supplementation with SFN. A secondary objective was to assess clinical response and adverse events (AEs) associated with supplementation with SFN alone.

| Study design and sample processing
On day 0 (D0), lymph node fine needle aspirates (FNA) were collected.
A total of 3 FNAs were obtained from at least 2 different peripheral lymph nodes from each dog. When possible, collection from mandibular lymph nodes was avoided. All dogs subsequently received 3 capsules of BroccoMax (Jarrow Formulas, Los Angeles, CA) twice daily by mouth for 7 days. BroccoMax is a commercially available broccoli seed supplement containing 30 mg of glucoraphanin per capsule, which yield approximately 8 mg of SFN after conversion by the enzyme myrosinase. 34 This dose was based on the bioavailability study performed by our research team in healthy dogs, 34 and on a similar dosing regimen utilized in a human breast cancer trial. 37 The period of supplementation was chosen based on several studies in mice and human suggesting that 1 to 7 days of supplementation was sufficient for the SFN to reach meaningful levels in various tissues, induce major proteomic changes in tumor cells, and impact the tumor burden as a whole in xenograft models. 11,16,20,38 Owners were directed to feed dogs their usual food, and avoid any treats for the duration of the trial. On day 7 (D7), duplicate lymph node samples were obtained after the last dose of supplementation with SFN. After sample collection on D7, dogs were deemed off-study and free to pursue mainstay chemotherapy treatment.
At D0 and D7, material obtained from the 3 FNA samples were mixed together at room temperature with 1 mL of 0.9% sterile saline in a cryovial. Immediately after, 400 μL of the lymph node samples and saline mixture were added, at room temperature, to another cryovial with 100 μL of RIPA protein lysis buffer (150 mM NaCl, 1% Triton X-100, 0.5% sodium deoxycholic acid [NaDOC], 0.1% SDS, 20 mM TRIS pH 8.0). The cryovial was then immediately flash frozen in liquid nitrogen. All cryovials were kept in liquid nitrogen for a maximum of 15 minutes before being stored at −80 C until further analysis.
All AEs occurring during supplementation with SFN were recorded by the owner, in a standardized questionnaire that was sent home on D0 with each dog. At D7, the AEs recorded were graded according to the VCOG common terminology criteria for AEs. 36 The clinical response to SFN at D7 was assessed via caliper measurement of peripheral lymph nodes. Disease response was assessed according to the VCOG response evaluation criteria. 39

| Immunoblots
The protein concentration of each sample was assessed using Pierce

| Gene set enrichment analysis
We used the R statistical computing environment (version 3.3.3) to generate ranked lists of genes from the protein abundance ratio data. All but 1 dog developed AEs, for a total of 24 episodes of AEs recorded. All AEs were grade 1 or 2, and most commonly lethargy, vomiting and anemia. More details are provided in Table 1 Homo sapiens database, 21 of these 24 proteins were precisely identified and the results are presented in Table 2.
A pitfall of retaining only proteins detected in over 50% of samples in each group is that proteins detected only at 1 time point but not at the other are excluded. Since these proteins could be biologically relevant, we further investigated them and displayed the results in Figure 1. We were able to identify 7 proteins that were detected only in the pre-SFN samples, and 21 proteins detected only in the post-SFN samples (Table 3).  Table 4. Of note, 7 of these proteins were also identified as significantly up-or down-regulated when all dogs were analyzed together. 3.4 | Mass spectrometry: Individual dogs at day 7 vs day 0

|
We then compared the differential expression of proteins before and after SFN administration in each dog. Of the 915 initial proteins, the number of proteins expressed at a level at least twice as low at D7 (FC D7/D0 < 0.5) or twice as high at D7 (FC D7/D0 > 2) ranged from 199 to 648 depending on the dog. As shown in

| Immunoblots
Two proteins were selected for validation of the LC-MS/MS by immunoblots: heat shock 70 interacting protein (HIP) and 14-3-3 protein theta (14-3-3-θ). As shown in Figure 2, the presence of HIP and

| Gene set enrichment analysis
We generated a ranked list of genes from the LC-MS/MS abundance ratio data as described in the Materials and Methods section. As some canine proteins were not encoded by a gene with a human ortholog, only 514 genes were included in the reference signature. Out of a  total of 5917 gene sets from the GO database, 5052 were filtered out after restricting to our data set. Therefore, a total of 865 gene sets were used in the analysis. The median number of genes per set was 21 (range, 10-155).
When only including gene sets with a P-value <.01 and a FDR qvalue <0.1, we identified 11 nonredundant gene sets (1.3%) significantly upregulated at D7. These gene sets included 2 main clusters: genes associated with immune response, and genes associated with protein maturation and transport, as detailed in Figure 4. Only 1 (0.1%) nonredundant gene set was significantly downregulated at D7. This gene set GO annotation was "tRNA processing" and the normalized enrichment score (NES) was −2.02 (P < .001).

| DISCUSSION
The study presented here investigates the impact of supplementation with SFN in cancer-bearing dogs. The results herein show that oral supplementation with twice daily SFN for 1 week was well tolerated in dogs with treatment-naïve multicentric LSA, and induced pronounced changes in the expression level of several hundred proteins.
Proteins and gene sets impacted by supplementation with SFN were frequently involved in regulation of innate and adaptive immunity, response to oxidative stress, gene transcription, apoptosis, and protein transport, maturation and ubiquitination.
Sulforaphane supplementation was well tolerated in this study.
Most dogs experienced AEs, but many were noted prior to the start of the study and the most severe events occurred in dogs that experienced LSA progression. Therefore, none of the AEs were considered likely related to SFN; however, SFN AEs causality is possible. No dog demonstrated a clinical response after a week of supplementation with SFN. Although SFN has a major impact on cancer cell growth and survival in vitro, [16][17][18][19][20]40 and slows the growth of certain tumor xenografts, 20,21 a measurable gross benefit to SFN-single-agent therapy in patients with naturally occurring cancer has not been reported.
When all dogs were considered, 24 proteins were significantly up-or down-regulated after supplementation with SFN. We found that, except for alanyl-tRNA synthetase, all these proteins have been studied in relation to the pathogenesis or prognosis of various human malignancies including renal, liver, colorectal, pancreatic, pulmonary, head and neck, breast, and ovarian cancer. [41][42][43] To our knowledge, none of these proteins have previously been investigated in veterinary oncology.
Thioredoxin related transmembrane protein 1 is involved in cellular response to oxidative stress and resistance to doxorubicin, cisplatin and etoposide in human T-cell leukemia cell lines. 44 We found this protein to be significantly downregulated after SFN (FC D7/D0 = 0.3).
Increased reactive oxygen species production is a key mechanism of SFN tumor suppression. In human prostate cancer cells, SFN led to apoptosis by depleting GSH levels and increasing oxidative stress. 9 The activity of thioredoxin and glutathione S-transferase in lung cancer cells lines was also inhibited by covalent binding of SFN to their cysteine residues. 23 To what extent alterations of cellular tolerance to T A B L E 5 Number of proteins upregulated and downregulated by a least 2 folds for each dog after 7 days of oral supplementation with sulforaphane Notes: Proteins with an FC D7/D0 < 0.5 were considered to be downregulated at D7. Proteins with an FC D7/D0 > 2 were considered to be upregulated at D7. F I G U R E 2 Immunoblots of lymph node samples for 7 dogs (d) with naïve multicentric lymphoma before (D0) and after (D7) supplementation with sulforaphane. The proteins HIP, A, and 14-3-3-θ, B, were detected in all samples (top rows). The immunoblots for β-actin (second rows) were used to normalize the band volume of HIP and 14-3-3-θ. The normalized expression ratio at D7/D0 is the ratio of the normalized band volume at D7 vs D0 for each dog (third rows). For comparison, the corresponding fold change at D7 vs D0 previously determined by LC-MS/MS is also annotated (bottom rows). FC D7/D0: Fold change at D7 vs D0 (mass spectrometry); NA: Not applicable; NER D7/D0: Normalized expression ratio at D7 vs D0 (immunoblot) F I G U R E 3 Enrichment plots of the most upregulated gene set (A, "regulation of protein maturation") and the most downregulated gene set (B, "tRNA processing") across all dogs postsulforaphane. The middle colored scale is a visual depiction of the ranked list of genes: the redder and more to the left, the most upregulated are the genes in the reference signature, the bluer and more to the right, the most downregulated are the genes in the reference signature. The vertical bars ("hits") immediately above the color scale indicate where the genes of the gene set appear in the reference signature. The top portion of the plot displays the running enrichment score for the gene set as the analysis walks down the ranked list of genes of the reference signature. The enrichment score for a gene set is the score furthest from zero (peak of the plot for an upregulated gene set, bottom of the plot for a downregulated gene set). The enrichment score reflects the degree to which the genes contained in a gene sets are overrepresented on 1 side or the other of the reference signature The proteins HIP and 14-3-3-θ were chosen to validate the mass spectrometry results as they were among those proteins exhibiting different expression between D0 and D7 according to the LC-MS/MS results. Additionally, the antibodies targeting these proteins were predicted to be cross-reactive with the canine protein according to the manufacturer. Based on LC-MS/MS, HIP was downregulated at D7 in dogs # 1, #3, #4 while it was upregulated in dogs #2, #5, #6.
These results were confirmed in all dogs via immunoblots. Moreover, 14-3-3-θ was downregulated at D7 in dogs #1, #2, #3, #4, and upregulated in dogs #5 and 6. The results were confirmed in 3 dogs via immunoblots, while for 2 dogs (#4 and #5), the LC-MS/MS and immunoblots results were reversed. The cause of the discordant F I G U R E 4 Bar graph displaying the normalized enrichment score of 11 gene sets significantly upregulated across all dogs at after sulforaphane. The gene sets, whose names appear on the left of each bar, are composed of genes annotated by the Gene Ontology (GO) term. Asterisks denote statistical significance as follows: *P < .01; **P < .001 results between the LC-MS/MS and immunoblots for 1 of the proteins in 2 dogs is unclear. The authors suspect a likely human error at 1 of the steps of the immunoblots for these dogs (protein quantification or dilution, electrophoresis, transfer on nitrocellulose membrane, antibody binding) since in our opinion, a human error is more likely to have occurred during these experiments than during mass spectrometry which is a more automatized and sensitive analysis.
To our knowledge, 5 proteomic studies of LSA in dogs have been published. [57][58][59][60][61] In 2 studies, the mass-to-charge ratio (m/z) of the protein peaks or the peptide mass profile were reported, however the identity of the proteins was not available. 58,61 In the other studies, the proteins differentially expressed between dogs with LSA and healthy dogs included haptoglobin, C-reactive protein, α2-macroglobulin, apolipoprotein A1 precursor, inter α-trypsin inhibitor, and several others. 57