Mammalian sphingosine kinase (SphK) isoenzymes and isoform expression: challenges for SphK as an oncotarget

Abstract

The various sphingosine kinase (SphK) isoenzymes (isozymes) and isoforms, key players in normal cellular physiology, are strongly implicated in cancer and other diseases. Mutations in SphKs, that may justify abnormal physiological function, have not been recorded. Nonetheless, there is a large and growing body of evidence demonstrating the contribution of gain or loss of function and the imbalance in the SphK/S1P rheostat to a plethora of pathological conditions including cancer, diabetes and inflammatory diseases. SphK is expressed as two isozymes SphK1 and SphK2, transcribed from genes located on different chromosomes and both isozymes catalyze the phosphorylation of sphingosine to S1P. Expression of each SphK isozyme produces alternately spliced isoforms. In recent years the importance of the contribution of SpK1 expression to treatment resistance in cancer has been highlighted and, additionally, differences in treatment outcome appear to also be dependent upon SphK isoform expression. This review focuses on an exciting emerging area of research involving SphKs functions, expression and subcellular localization, highlighting the complexity of targeting SphK in cancer and also comorbid diseases. This review also covers the SphK isoenzymes and isoforms from a historical perspective, from their first discovery in murine species and then in humans, their role(s) in normal cellular function and in disease processes, to advancement of SphK as an oncotarget.

Keywords: cancer; isoenzymes; isoforms; sphingosine kinase; sphingosine-1-phosphate.

Figure 1. The sphingosine kinase (SphK) rheostat in the maintenance of cell proliferation and function

SphK is a phospholipid enzyme that converts sphingosine to S1P. Through the repression of sphingosine by phosphorylation, SphK modulates the balance between S1P and sphingosine and ceramides to promote cell survival, normal cell proliferation and cell function. SphK has intrinsic catalytic activity to facilitate a housekeeping role in maintaining physiological levels of sphingosine and ceramide and is also stimulated by mitogens such as growth factors, estradiol, and ERK. On the other hand, S1P lyase irreversibly degrades S1P into hexadecenal and phosphoethanolamine while S1P phosphatase dephosphorylates S1P to sphingosine to maintain physiological levels of S1P and maintain homeostasis.

Figure 2. Subcellular distribution of SphK isozymes and function

A. SphK is expressed as two isozymes designated SphK1 and SphK2. Each SphK isozyme has variant isoforms differing only at the N-terminus. B. The distinct functions of hSphK1 and hSphK2 isozymes are believed to be associated with subcellular localization. SphK2 isoforms are predominantly localized in the nucleus, mitochondria, and endoplasmic reticulum (ER) and produce S1P. While located at the ER, SphK phosphorylates sphingosine to S1P, conversely, S1P phosphatase removes the phosphate and S1P lyase degrades S1P to induce apoptosis. In the mitochondria, Sphk2 catalysis to produce S1P triggers the apoptotic pathway by activating BAK and Cyt C. Conversely, hSphK1 is predominantly located in the cytoplasm and translocates to the membrane upon activation. Upon activation, hSphK1a/b/c phosphorylates sphingosine to produce S1P and has an “inside/outside” mechanism whereby S1P is translocated outside the cell and can bind to S1P_1-5 receptors (also referred to as G-coupled receptors) contributing mainly to cell survival. HSphK1 isoforms have also been found extracellular. In mice, SphK1 isoform b has been shown to be located to the plasma membrane and more susceptible to degradation, however, to-date, no distinction between hSphK1 isoform localization has been demonstrated in the different human cells and human tissue with the exception that hSphK1a has been shown to be distributed in the nucleus and cytoplasm in cell culture. Figure 2B has been adapted from [27].

Figure 3. Designer anti-S1PR1-5 drug specificity

Update of 1^st and 2^nd generation S1P receptor modulators developed to target individual and multiple S1P receptors. Each of the receptor modulators binds to individual or multiple receptors to block or activate the SIPRs as illustrated. MOST S1PR modulators target 2 or more receptors. Sipon (BAF312) - Siponimod is a S1P1 = 5 > 4 modulator; Ponesimod (ACT-128800) is an agonist for S1P1 > 5 > 3; KRP-203 is an agonist for S1P1 > 4; Ceralifimod (ONO-4641) is an agonist for S1P1 = 5 > 4; Ozanimod (RPC1063) is an agonist for S1P1 > 5 > 4; Cs-0777 is an agonist for S1P1 > 5 > 3; GSK2018682 is an agonist for S1P1 > 5. JTE-013 is a competitive antagonist specifically for S1P2. CYM50367 targets S1PR1/4 and VPC23019 targets S1PR1/3. Sew2871, AUY954, MT-1303, CYM-5442, W146, NIBR-0213, TAS0277308, are listed as selective S1P1 modulators. These novel S1P receptors and downstream signaling pathways and functions are reviewed in [15, 258, 281].

Figure 4. Alignment of the N-terminal amino acid sequences of the mSphK1 and hSphK1 isoforms

A. Three mSphK1 isoforms (iso-a, iso-a2, and iso-b) have been identified to date. Iso-a has an additional 6 and 7 amino acids at the N-terminus (N’) compared to the mSphK1a2 and mSphK1b isoforms respectively. There is 10 amino acid overall sequence dissimilarity when comparing mSphK1a to mSphK1a2 and mSphK1b. MSphK1a2 differs from mSphK1b by an insertion of a valine (V) in the N’ changing the sequence MEPVE (iso-2a) to MEPEE (iso-b). B. Human SphK1 is expressed as 3 isoforms (a, b, and c) which have identical in sequences except for the N-terminus (N’). The SphK1b N’ has an additional 86 amino acids upstream of the Met start codon of SphK1a, as illustrated. SphK1c has an additional 14 amino acids upstream from the SphK1a start codon, which translates to a difference of 17 amino acids between the SphK1a isoform compared to SphK1c. Legend: * = complete amino acid sequence identity across all isoforms; : = similarity of amino acids across 2 isoforms. Sequences aligned using CLUSTAL Omega (1.2.3) multiple sequence alignment tool.

Figure 5. Comparative alignment of human and murine SphK1 amino acid sequences

SphK1 amino acid sequence alignment showing similarities and differences in SphK1 sequences across human (H. sapiens), mouse (Mus musculus) and rat (R. norvegicus) species. Legend: * = complete amino acid sequence identity across all three species; : = similarity of amino acids across 2 species. Sequences aligned using CLUSTAL Omega (1.2.3) multiple sequence alignment tool.

Figure 6. Comparative alignment of human and murine SphK2 amino acid sequences

SphK2 sequences have been aligned to show similarities and differences between the SphK2 isozyme sequences across human (H. sapiens), mouse (Mus musculus) and rat (R. norvegicus) species. Legend: * = complete amino acid sequence identity across all three species, : = similarity of amino acids across 2 species. Sequences aligned using CLUSTAL Omega (1.2.3) multiple sequence alignment tool.

Figure 7. Predicted N-terminal sequence alignment of the hSphK2 isoforms

hSphK2 has 5 predicted isoforms differing only at the N-terminus as shown (iso-a, iso-b, iso-c and iso-d). The SphK2c isoform has a predicted 59 amino acid insertion in the N-terminal region compared to SphK2b. SphK2c has a predicted Met start site that is 36 amino acids downstream of the SphK2a (iso-a) start codon. SphK2d has the shortest sequence with a truncation of 86 amino acids from the SphK2a start codon. Legend: * = complete amino acid sequence identity across all isoforms. Sequences aligned using CLUSTAL Omega (1.2.3) multiple sequence alignment tool.