ABBSIN_Contents 1..1 - CiteSeerX

0 downloads 0 Views 1006KB Size Report
Nov 18, 2008 - (breast muscle and gastrocnemius) than in white ones, composed of glycolytic fibers (biceps brachii). We had found this observation very ...
Acta Biochim Biophys Sin (2009): 280 – 284 | ª The Author 2009. Published by ABBS Editorial Office in association with Oxford University Press on behalf of the Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences. DOI: 10.1093/abbs/gmp011. Advance Access Publication 8 March 2009

Expression of cytosolic 50 nucleotidase does not correlate with expression of oxidative metabolism marker: myoglobine in human skeletal muscles Katarzyna Lechward* and Kinga Tkacz-Stachowska Department of Molecular Enzymology, Intercollegiate Faculty of Biotechnology, Medical University of Gdansk, Gdansk, Poland *Correspondence address. Tel: þ48-58-3491470; Fax: þ48-58-3491445; E-mail: [email protected]

Our previous studies had shown that cytosolic 50 nucleotidase-I (cN-I) is expressed in several tissues in pigeons, including brain and several different skeletal muscles. We observed that cN-I mRNA levels varied among different pigeon muscles. Initial quantification of the differences revealed that 5 –10 times more of cN-I transcript was present in red, oxidative muscles (breast muscle and gastrocnemius) than in white ones, composed of glycolytic fibers (biceps brachii). We had found this observation very intriguing and decided to compare human skeletal muscles distribution of cN-I with the type of oxygen metabolism. Our screen involved 60 samples of several human muscles and we assayed the correlation between the amount of transcripts of cN-I and myoglobine, which we took as a measure of oxidative-slow twitch fibers. Our question was whether in humans, cN-I presence in skeletal muscles was related to their fiber composition. If that was the case, then cN-I expression could serve as a tool to assess the percentage of oxidative fibers in any given human muscle sample, where myoglobine expression could not be readily measured. After quantification of expression of both genes, we concluded that there was no correlation between expression of cN-I and fiber type. Therefore, contrary to the pigeon muscles, cN-I did not reflect the ratio of oxidative fibers to the total mass of the muscle sample in humans. That difference indicated that there were certain mechanisms that differentially regulated the expression of cN-I in muscle tissues of mammals and lower vertebrates.

Keywords ATP metabolism; cytosolic nucleotidase; adenosine; human skeletal muscle Received: November 18, 2008

Accepted: February 11, 2009

Acta Biochim Biophys Sin (2009) | Volume 41 | Issue 4 | Page 280

50

Introduction ATP metabolism is crucial for energetic homeostasis of every living cell and its balancing is of primary importance of surviving and maintaining the sense of wellbeing [1,2]. Several mechanisms based on positive and negative feedback loops are implicated into regulation of this balance, including a key player adenosine. Adenosine is produced by the group of enzymes called 50 nucleotidases (50 NTs). 50 NTs are widely expressed from bacteria to humans. They are classified into several groups, depending on the cellular localization and substrate specificity [3,4]. In humans, there are three types of cytosolic 50 nucleotidases: type I with A and B isoform, type II, and type III. The cytosolic 50 nucleotidase-I (cN-I) A protein is ubiquitously expressed, whereas the B isoform was never found as a protein but its RNA was highly expressed in testis. cN-I type proteins prefer AMP as substrate contrary to 50 cytoplasmic nucleotidase type II, which prefers IMP over AMP. Type III has high affinity to UMP and CMP. Apart from soluble enzymes, there are membrane-bound 50 NTs, so-called ecto-50 NTs, which are mainly engaged in the cell adhesion, and their amount rises in several types of cancers, including breast and colon cancers. Our interest mainly focuses on soluble, cN-I protein, and we have cloned, biochemically characterized, and determined its expression in a panel of pigeon tissues [5]. We repeatedly observed that in so-called red, slow-twitch, oxidative muscles, the expression of both cN-I RNA and protein was noticeably higher than in so-called fast-twitch, white glycolytic ones. On the basis of that observation, we hypothesized that there could be a correlation between expression of cN-I and the type of

Cytosolic 50 nucleotidase expression in human skeletal muscle

metabolism in different muscles. We decided to put this hypothesis under the tests and screen human muscle samples for the expression of cN-I type A and B as well as marker of oxidative metabolism-myoglobine [6].

Materials and Methods Chemicals Ammonium persulfate, Ficoll 400, bromophenol blue, and sodium citrate were from SERVA Electrophoresis (Heidelberg, Germany). Protein inhibitors and CDP-star were purchased from Roche Molecular Biochemicals (Mannheim, Germany). The reverse transcription system and all enzymes for PCR were supplied by Fermentas (Opelstrasse, Lithuania) and molecular weight markers for SDS–PAGE were purchased from Pharmacia (Stockholm, Sweden). All other reagents were of the highest analytical grade available. Human muscle samples Fragments of human muscles were obtained from the Department of Emergency Surgery at Medical School of Gdansk, Poland. Muscle samples (0.5–2 g) were taken during routine total hip replacement surgery and from emergency cases. The fragments were frozen in liquid nitrogen immediately after removal and each sample was subsequently portioned for northern blot, western blot, and RT –PCR experiments. We had collected 45 samples from muscle quadriceps femoris from senior group of patients, and the 15 samples originated from junior groups: muscle gluteus maximus [1], m. quadriceps femoris [1], muscle tibialis anterior [4], muscle brachoradialis [3], muscle deltoideus [3], and muscle biceps brachii [3]. Northern blot analysis Total cellular RNA was isolated from muscle samples by the method of Chomczynski and Sacchi [7] based on acid guanidinium thiocyanate –phenol–chloroform extraction. Northern blot analysis was routinely performed on 10–25 mg of total RNA as described previously [5]. Intensities of signals were quantified by Sigma Scan software and statistical analysis was performed with Statistica 6.0 software. SDS– PAGE and western blot analysis Human muscle fragment homogenates were prepared as described previously [8], and as standard we assayed 150 mg of protein on SDS –PAGE followed by transfer onto nitrocellulose membrane. Dilutions of the antibodies were as follows: 1:100 anti-cN-I (kindly provided

by Graciela Sala-Newby from Bristol Heart Institute, UK), and 1:500 anti-cN-I A and 1:500 anti-cN-I B (both provided by Jozef Spychala from North Carolina Cancer Institute, Chapel Hill, NC, USA). Immune complexes were visualized by the ECL system. The 47 kDa human recombinant cN-I protein provided by Jozef Spychala (North Carolina Cancer Institute) served as a positive control in all experiments.

Results Classification of muscle samples according to cN-I gene and protein expression To verify expression levels of cN-I, total RNA was isolated from all samples according to the method described earlier. The procedure yielded between 42 and 231 mg RNA with appropriate OD260/280 value (1.7 –2.1). In the first approach of the study, we classified the muscle samples according to the expression levels of cN-I gene and cN-I protein by northern and western blot techniques, respectively. Six types of muscles were chosen: m. gluteus maximus, m. quadriceps femoris, m. tibialis anterior, m. brachoradialis, m. deltoideus, and m. biceps brachii. All muscle samples, except for m. quadriceps femoris, originated from 10 healthy men of the similar age (22–31 years), living active, sporty lifestyle, and undergoing the bone fracture-related treatments. The number of patients is smaller than the number of samples collected due to several cases of multiple limbs fractures. The patients were subjected to surgery due to different injuries such as serious limbs breakage, internal fixation, or replacement of fracture. m. quadriceps femoris was taken from a 38-year-old man during total hip replacement operation. Results were confirmed by running each RNA sample at least twice and performing hybridization procedures under the same experimental conditions. Figure 1 shows the representative northern blot analysis of RNA isolated from m. gluteus maximus, m. quadriceps femoris, m. tibialis anterior, m. brachoradialis, m. deltoideus, and m. biceps brachii, with 18S RNA serving as loading control. Human cN-I gene expression was found to vary in different muscles, with the highest in m. quadriceps femoris and m. gluteus maximus and the lowest in m. deltoideus and m. brachoradialis. Our next question was whether the levels of RNA encoding cN-I corresponded to actual amount of protein present in the tissues, as one could expect different posttranscriptional control mechanisms operating there. Therefore, we performed immunoblotting on high-speed cytoplasmic muscle extracts with antibodies against Acta Biochim Biophys Sin (2009) | Volume 41 | Issue 4 | Page 281

Cytosolic 50 nucleotidase expression in human skeletal muscle

Fig. 1 Northern blot analysis reveals cN-I RNA distribution in human muscles RNA samples purified from various muscle tissues were resolved, blotted, and probed with radioactive-labeled probes against cN-I and 18S RNA. Autoradiogram shown here represents one out of three experiments performed on the same RNA samples.

human cN-I (Fig. 2) and human cN-I A and B isoformspecific antibodies (Fig. 3). As shown in Fig. 2, antihuman cN-I antibodies recognized major band of 46 kDa. Two smaller and less intense bands of a mass of 42 and 30 kDa were most likely to be proteolytic fragments on cN-I. Antigen competition experiments proved that all of the bands were specific (data not shown). When one compared the levels of RNA and protein for single sample, it can be seen that the amount of RNA is proportional to the amount of proteins. Using the recombinant cN-I protein of the known mass (47 kDa) as a standard, the molecular weight of human muscle cN-I could be estimated at the same level. The relative amounts of protein correlated with observed mRNA levels in every skeletal muscle sample tested, were the most significant in m. quadriceps femoris and m. gluteus maximus and the lowest in m. deltoideus and m. brachoradialis. Major conclusion taken from Fig. 3(A) was that the isoform expressed in tested samples was cN-I A. We

Fig. 2 cN-I protein expression in human muscles Cytoplasmic fractions of different skeletal muscle were resolved on SDS– PAGE and blotted against cN-I protein. St referred to recombinant human cN-I. Antibodies used in this experiment did not discriminate between cN-I A and cN-I B isoforms. The samples represented individuals not pools of all collected material from the same anatomical region. Acta Biochim Biophys Sin (2009) | Volume 41 | Issue 4 | Page 282

Fig. 3 Only cN-I A protein was expressed in human skeletal muscles Cytoplasmic extracts from three different muscles were SDS –PAGE resolved, transferred onto nitrocellulose membrane, and stained with anti-cN-I A (A) and anti-cN-I B (B) polyclonal antibodies, as stated in the Material and methods section.

tried to assay the expression of cN-I B isoform by both, Western blot analysis [Fig. 3(B)] and RT–PCR approach (data not shown), but we obtained no signals from any of the samples tested. The protein of 47 kDa was detected only with hcN-I A antibodies in all three muscle samples. This emphasizes the suggestion that the only cN-I isoform expressed in human skeletal muscles is hcN-I A. Since the prevailing type of muscle tissue which we obtained was m. quadriceps femoris, we focused on those samples and decided to do more detailed statistical analysis of correlation between cN-I expression and oxidative metabolism.

Classification of m.quadriceps femoris muscle samples according to myoglobine gene expression The level of oxidative metabolism in 38 samples of m. quadriceps femoris was assessed by quantifying the amount of myoglobine transcript. Myoglobine, despite some controversies found in the literature, was one of the most relevant markers of oxidative fibers and we took it as a reference for our screen. We assayed muscle samples that originated from highly homogenous group of patients in terms of disability (total hip replacement), age (69–74), and gender (slight prevailing men: 57.9%). Taking into account the type of performed surgery and the age of patients, we assumed that their physical activity was rather low with the periods of immobilization. Tested muscle samples showed significant variation in respect of myoglobine message levels. We had quantified the intensities of signals by means of Sigma Scan software and standardized it with the expression of 18S RNA gene. Based on that quantification, we created

Cytosolic 50 nucleotidase expression in human skeletal muscle

Table 1 Myoglobine expression in human quadriceps femoris muscle samples

Expression level of myoglobine gene

Value

Sample (n)

Percentage (%)

2 þ þþ þ þþ

,3271 3272 – 6541 6542 – 9813 9814 – 13084

3 19 11 5

7.9 50.0 29.0 13.1

Table 2 cN-I expression levels in human quadriceps femoris muscle samples

Expression level of cN-I gene

Value

Sample (n)

Percentage (%)

2 þ þþ þ þþ

,3271 3272 – 6541 6542 – 9813 .9813

5 13 17 3

13.1 34.2 44.8 7.9

four groups characterized by none (2), moderate (þ), high (þþ), and very high (þþþ) myoglobine (Table 1) and cN-I (Table 2) genes expression in 38 quadriceps femoris muscle fragments from a group of patients of different sex (57.9% of men) in the age ranging from 69 to 74 years. The same set of data was used in Fig. 4, yet Tables 1 and 2 show how the values were distributed in the population, whereas Fig. 4 delineates what is the real relation between the measured values and ideal correlation curve. Figure 4 shows the

correlation plot between myoglobine and cN-I gene expression levels. The P-value was calculated as 0.508, what showed that there was no relationship between expression levels of cN-I and myoglobine in human muscles. There was little doubt that cN-I was expressed independently of the level of oxidative metabolism in 38 human quadriceps femoris muscle fragments.

Discussion To our knowledge, this is the first attempt to correlate and quantify the expression of cN-I, engaged in ATP catabolism with fiber composition of human skeletal muscles. Our studies showed that the predominant form of cytosolic 50 nucleotidase, which resides in human muscles, was cN-I A isoform, and that the B isoform was not expressed on RNA level or on protein level. What we were interested in was to understand whether there was any correlation between cN-I expression and myoglobine expression, in the samples of muscles originated from two different age groups: so-called juniors (,35 years old) and seniors (.60 years old). There are several research papers that link adenosine content, age, and physiological activity with fiber composure of the muscles in rodents, but so far the exact molecular mechanisms underlying observed dependencies and correlations remain elusive [8 –11]. We found some pieces of evidence that showed better correlation of oxidative metabolism and cN-I expression in muscle samples obtained from young, sporty, and healthy men under 30 years old, but the limited availability of the material did not allow us to perform extensive statistical analysis to

Fig. 4 Expression of myoglobine and cN-I RNA values was plotted against each other and the correlation was assessed After quantification of signals obtained from northern blot, correlation analysis was performed with Statistica 6.0 software. It determined P-value to be 0.508 for 38 samples tested. Acta Biochim Biophys Sin (2009) | Volume 41 | Issue 4 | Page 283

Cytosolic 50 nucleotidase expression in human skeletal muscle

confirm initial results, yet P-value correlation factor calculated for this group reached 0.7. More over, the amount of adenosine 50 NT depended on the type of muscle, in junior group, whereas in the samples from older patients, these differences were less pronounced. Our study applied more detailed analysis of expression of ATP catabolism-related enzymes with parameters listed above and in consequence reveal all the links among glycolysis, oxidative phosphorylation, and adenosine metabolism in muscle tissues. We concluded that in mammalian muscles, mechanisms that couple expression of genes responsible for adenosine catabolism with metabolic processes were certainly different from those present in avias, represented by pigeon breasts muscles. In contrast to heart tissue, where the adenosine participates in ischemic preconditioning, human skeletal muscles are not expected to be protected by it in the same way, yet adenosine must play a role in the regulation of the strength and time of the contractions, during the work cycle. Following the initial observation about the uneven distribution of cN-I gene and protein in fast- and slow-twitch muscles, we undertook the work in humans, but we did not incorporate so-called behavioral and physiological differences between both species. The major locomotion mechanism utilized by pigeons is an active flight, and they cross very short distances by walking, when compared with humans whose main way of locomotion is walking. Humans, in general, use the whole array of skeletal muscle to keep the vertical posture during locomotion, yet they have very diverse life style, which includes different levels of daily exercises, the walked distances, and the usage of mechanical engine-based means of transportation tools in the urban zones.

Acknowledgements We would like to thank our colleagues, surgeons from the Department of Emergency Surgery Medical University of Gdansk, for excellent help with collecting material presented in this study, coworkers from the Intercollegiate Faculty of Biotechnology of UG-MUG for technical support and discussions, and Professors Wieslaw Makarewicz and Andrzej Cezary Skladanowski

Acta Biochim Biophys Sin (2009) | Volume 41 | Issue 4 | Page 284

for providing us with scientific freedom to perform experiments presented above.

Funding Molecular biology reagents including enzymes, except Klenow polymerase were purchased from grant from Polish State Committee for Research no PO406015 for KL, and the rest of reagents were purchased from overhead costs of the Laboratory of Molecular Enzymology. Klenow polymerase was kind gift of Dr Joanna Jakubkiewicz-Banecka, from Department of Molecular Biology of University of Gdansk.

References 1 Bianchi V and Spychala J. Mammalian 50 nucleotidases. J Biol Chem 2003, 278: 46195– 46198. 2 Hellsten Y and Frandsen U. Adenosine formation in contracting primary muscle cells and endothelial cells in culture. J Physiol 1997, 504: 695– 704. 3 Hunsucker SA, Spychala J and Mitchell BS. Human cytosolic nucleotidase I. Characterisation and role in nucleoside analog resistance. J Biol Chem 2001, 276: 10498– 10504. 4 Sala-Newby GB, Skladanowski AC and Newby AC. The mechanism of adenosine formation in cells. Cloning of cytosolic 50 nucleotidase-I. J Biol Chem 1999, 274: 17789 –17793. 5 Tkacz-Stachowska K, Lechward K and Skladanowski AC. Isolation and characterization of pigeon breast muscle cytosolic 50 nucleotidase-I (cN-I). Acta Biochim Pol 2005, 52: 789 – 796. 6 Brunori M. Nitric oxide, cytochrome c-oxidase and myoglobine. Trends Biochem Sci 2001, 26: 21 –23. 7 Chomczynski P and Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987, 162: 156– 159. 8 Brault JB and Terjung RL. Purine salvage to adenine nucleotides in different skeletal muscle fiber types. J Appl Physiol 2001, 91: 231– 238. 9 Brotto MA, Nosek TM and Kolbeck RC. Influence of ageing and the fatigability of isolated skeletal muscles from mature and aged mice. Exp Physiol 2002, 87: 77 – 82. 10 Cheng B, Essackjee HC and Ballard HJ. Evidence for control of adenosine metabolism in rat oxidative skeletal muscle by changes in pH. J Physiol 2000, 3: 467– 477. 11 Contley KE, Blei ML, Richards TL, Kushmerick MJ and Jubrias SA. Activation of glycolysis in human muscle in vivo. Am J Physiol 1997, 273: C306 –C315.