Comparative time-dependent analysis of potential inflammation biomarkers in lymphoma-bearing SJL mice

Abstract

SJL mice colonized with RcsX lymphoma cells undergo a rapid inflammatory response associated with biological and physiological effects including increased nitric oxide production and mutations in spleen DNA. By 2 weeks postcolonization, these changes were accompanied by both up- and down-regulation of a number of plasma proteins. In the experiments reported here, plasma from individual SJL mice was analyzed at several time-points over the 2-week period to determine if there were sets of proteins whose expression varied in concert and thus might serve as early biomarkers for inflammation-related disorders. Samples were collected just prior to injection of the RcsX cells and then after 4, 8, and 12 days. Albumin and immunoglobulins were depleted, and the samples were resolved by 1D gel electrophoresis. The gels were cut into 20 slices, and the proteins were digested in-gel with trypsin. The digests were treated with iTRAQ reagents and then analyzed using LC/MS/MS. The resulting data were processed with two software packages, that is, ProQuant and Spectrum Mill, and then subjected to K-means cluster analysis (K = 4). The four clusters revealed a set of highly up-regulated proteins, a set of progressively up-regulated proteins, a set with no major changes, and a set that declined. The first cluster included haptoglobin and serum amyloid A; the second included groups with several functions including protease inhibition, cell motility, and transport. The iTRAQ results for a selection of the up-regulated proteins, including haptoglobin, hemopexin, serum amyloid P component, and ceruloplasmin, were confirmed with Western blots. Prominent down-regulated proteins included esterase-1, paraoxonase, and alpha-2-macroglobulin. Approximately 50% of the up-regulated proteins are canonical acute phase proteins, while the remainder are regulated by the Nrf2 transcription factor.

Figure 1

Schematic summary of the experimental design. Animals were injected with RcsX tumor cells on day 0, and plasma sampled on days 0,4, 8, and 12. Samples from mice 1 and 2, and 3 and 4, were combined; these samples and those from mouse 5 were analyzed by 1D gel electrophoresis. The gels were cut into twenty equal slices and slices from the mid range were combined as shown, reduced, alkylated, and digested in-gel. Samples from days 0, 4, 8, and 12 were labeled with iTRAQ reporters 114, 115, 116, and 117, respectively, and analyzed by LC/MS/MS.

Figure 2

A Venn diagram showing the number of proteins identified by Spectrum Mill in individual sets of samples. A total of 157 proteins were identified from the three sets of mouse plasma in which at least a 2-fold difference was observed in the level of 42 proteins in at least one time point as listed in Table 3.

Figure 3

K-means cluster analysis of 72 proteins that were found in two separate sets of samples and identified with both ProQuant and Spectrum Mill. Proteins were clustered using the expression levels at the three time points as variables. K was set to 4. Means of the natural logged ratios for each cluster group and time point are shown with error bars representing the standard deviation.

Figure 4

Quantitative immunoblot analysis for five representative proteins identified by iTRAQ analysis: complement component 4 (B), hemopexin (C), ceruloplasmin (D), serum amyloid P (E), and haptoglobin (F) in mouse plasma during tumor growth in SJL mice. Figure 4A: Plasma proteins (5μg) from two sets of SJL mice were collected at 0, 4, 8, and 12 days after RcsX-tumor cell injection, separated by SDSPAGE, then transferred to PVDF membranes and probed with the indicated antibodies. Figure 4B-F: Relative ratios with respect to day 0 for complement component 4 (B), hemopexin (C), ceruloplasmin (D), haptoglobin (E), and serum amyloid P (F) using ProQuant.