Calreticulin expression in the clonal plasma cells of patients with systemic light-chain (AL-) amyloidosis is associated with response to high-dose melphalan

Abstract

In high doses with stem-cell transplantation, melphalan is an effective but toxic therapy for patients with systemic light-chain (AL-) amyloidosis, a protein deposition and monoclonal plasma cell disease. Melphalan can eliminate the indolent clonal plasma cells that cause the disease, an achievement called a complete response. Such a response is usually associated with extended survival, while no response (a less than 50% reduction) is not. Gene-expression studies and a stringently supervised analysis identified calreticulin as having significantly higher expression in the pretreatment plasma cells of patients with systemic AL-amyloidosis who then had a complete response to high-dose melphalan. Calreticulin is a pleiotropic calcium-binding protein found in the endoplasmic reticulum and the nucleus whose overexpression is associated with increased sensitivity to apoptotic stimuli. Real-time PCR and immunohistochemical staining also showed that expression of calreticulin was higher in the plasma cells of those with a complete response. Furthermore, wild-type murine embryonic fibroblasts were significantly more sensitive to melphalan than calreticulin knock-out murine embryonic fibroblasts. These data have important implications for understanding the activity of melphalan in plasma-cell diseases and support further investigation of calreticulin and its modulation in patients with systemic AL-amyloidosis receiving high-dose melphalan.

Figure 1

Clonal plasma cells from patients with systemic AL-amyloidosis can be highly enriched after FACS-sorting. (A) Flow cytometric plot of FACS-sorted cells is shown with CD138⁺/DAPI⁻ cells in the bottom right quadrant. CD138⁺/DAPI⁻ cells were used for gene-expression analysis in 9 cases. In the insert, an IFM image of enriched clonal plasma cells from the marrow aspirate of an AL-amyloidosis patient is shown. These cells were stained intracellularly for lambda light chains with a FITC-linked monoclonal antibody and were more than 95% pure. (B) Box-and-whiskers plots of the expression levels of lineage-specific genes are depicted for the 9 specimens of purified clonal plasma cells used in gene-expression studies. The box extends from the 25th percentile to the 75th percentile, with a horizontal line at the median (50th percentile), and the whiskers extend down to the smallest and up to the largest value. The genes (and probe sets in parentheses) include genes for clonality, namely the λ and κ light-chain constant region genes (Affymetrix probe sets 221651 and 215121). Eight of the 9 specimens were λ and hence the nonclonal gene was κ in those cases. The plot of expression levels for the clonal and nonclonal genes clearly shows that the clonal genes were expressed at higher levels than the nonclonal constant region genes (paired t test, P = .005). The other lineage markers were CD38 (205692), XBP-1 (200670), CD138 (201286), CD4 (200670), CD14 (201743), CD16 (206398), CD19 (206398), CD33 (206120), CD45 (212588) and CD64 (214511). Myeloma cells can aberrantly express some of these markers. Comparison of the CR and NR sets showed no significant differences in gene- expression levels by paired t test for clonal and nonclonal IgV_L constant region genes, CD38, XBP-1, CD138, or other lineage markers with expression levels more than 100.

Figure 2

Expression levels of calreticulin by real-time (qRT-) PCR correlate with response to high-dose melphalan. Twenty-four specimens of purified plasma cells from patients with systemic AL-amyloidosis were obtained before treatment with high-dose melphalan and assessed for calreticulin expression by qRT-PCR. GAPDH and pooled cDNA from the purified plasma cells of 3 patients with hereditary amyloidosis were used as controls. Patients were assessed for response to high-dose melphalan at 3 months after treatment. Calreticulin expression in responders was significantly higher than in nonresponders (P < .01, Mann-Whitney). The box extends from the 25th percentile to the 75th percentile, with a horizontal line at the median (50th percentile), and the whiskers extend down to the smallest and up to the largest value.

Figure 3

Calreticulin expression is variable in the plasma cells of patients with systemic AL-amyloidosis by immunohistochemical staining of bone marrow biopsies. In these images from adjacent sections of bone marrow biopsies from 2 patients, CD138⁺ plasma cells stained immunohistochemically are shown on the left and calreticulin on the right. A pathologist who did not know the clinical status of the patients scored all cases blindly for intensity and distribution of calreticulin to compute the calreticulin index. In panel A, the calreticulin index of 9 is a function of the intensity of staining, indicating that calreticulin was found in a pancytoplasmic distribution. In panel B, the calreticulin index of 3 is a function of an intensity score of 1, indicating that the distribution of calreticulin was rare and scattered.

Figure 4

The distribution of calreticulin indices and response to high-dose melphalan. Immunohistochemical staining of baseline marrow biopsies was performed as described. (A) Distribution of calreticulin indices for patients with CR (n = 20), PR (n = 12), and NR (n = 16) shown. The difference between CR and NR groups is significant and there is a trend toward significance in the comparison of CR and PR groups. (B) qRT-PCR levels of patients with indices of 1 to 4 (n = 7) are compared with those of patients with indices 6 to 9 (n = 17; P = .09, unpaired t test). Horizontal bars are mean values.

Figure 5

The response of wild-type and calreticulin knock-out murine embryonic fibroblasts (MEF) to melphalan exposure. (A) Wild-type and calreticulin knock-out MEF cells were seeded onto flat-bottom 2 mL wells at 10⁵ cells per well in standard medium overnight and were then exposed to increasing doses of melphalan for 24 hours. Cells were then harvested with 0.1% trypsin and counted for viability with trypan blue staining. The results of 3 experiments are shown. The wild-type MEF cells (—●—) are significantly more sensitive to all doses of melphalan than the calreticulin knock-out cells (—○—; mean ± SD; P= .04, 2-tailed paired t test). (B) Lactate dehydrogenase release was directly measured to assess the cytotoxic effects of increasing doses of melphalan on wild type and calreticulin knock-out MEF cells. Triton-X 100–treated wells provided positive controls. The results of 3 experiments are shown. The wild-type MEF cells (—●—) are significantly more sensitive to all doses of melphalan than the calreticulin knock-out cells (—○—; mean ± SD; P = .006, 2-tailed paired t test).