INTRODUCTION: Bladder contractility in the transverse direction is often overlooked, because longitudinal strips are the regular tissues of choice in most contractility studies. In the present study, the effects of K+ channel blockers on transverse and longitudinal rat detrusor contractility were compared.
METHODS: Detrusor strips in transverse and longitudinal directions were dissected from young adult rats. Isometric tension was monitored using a myograph. The effects of tetraethylammonium chloride (TEA), 4-aminopyridine (4-AP), glibenclamide (Glib), iberiotoxin (IbTX), charybdotoxin (ChTX), and apamin on carbachol (CCh)-induced contractions were examined.
RESULTS: No contractile differences were present between transverse and longitudinal strips following CCh stimulation. Equal sensitivity to 4-AP and IbTX was detected in transverse and longitudinal strips. Pretreatment with Glib or ChTX resulted in greater suppression of CCh contractions in longitudinal strips. Although apamin suppressed contractions in both transverse and longitudinal strips, CCh potency was lower in transverse strips only.
CONCLUSION: Functional heterogeneity of transverse and longitudinal detrusor contractility was revealed from selective K+ channel blockade. Longitudinal strips were more susceptible to ATP-sensitive and intermediate-conductance Ca2+-activated K+ channel blockades, whereas transverse strips were affected more by blockade of small-conductance Ca2+-activated K+ channels. The potential importance in evaluating multidirectional contractility in pharmacologic studies of detrusor smooth muscle is reinstated.
KEYWORDS: K+ channel; Rat detrusor; Contraction; Transverse; Longitudinal
CORRESPONDENCE: Willmann Liang, PhD, School of Biological Sciences, Nanyang Technological University, Singapore 637551 ().
CITATION: Urotoday Int J. 2010 Apr;3(2). doi:10.3834/uij.1944-5784.2010.04.07
ABBREVIATIONS AND ACRONYMS: 4-AP, 4-aminopyridine; BK, large conductance Ca2+-activated K+; CCh, carbachol; ChTX, charybdotoxin; CRC, concentration-response curve; Emax, maximal fitted response; Glib, glibenclamide; IbTX, iberiotoxin; IK, intermediate conductance Ca2+-activated K+; SK, small conductance Ca2+-activated K+; TEA, tetraethylammonium chloride
Urine storage and release are primary functions of the urinary bladder, made possible by smooth muscle relaxation and contraction, respectively. The detrusor muscle of the bladder consists of a vast number of smooth muscle cells . Organization of smooth muscle bundles in the detrusor does not follow a distinguished pattern. Instead, cells run in different directions within the bladder wall. In order to expel and store urine, the bladder (in particular the detrusor) must contract and relax circumferentially.
Developed tension in the longitudinal direction has previously been used as a parameter of measurement in isolated bladder studies investigating both unstimulated and drug-modulated contractions. However, it remains questionable whether longitudinal contractility alone is sufficiently representative of the contractile characteristics of whole-bladder mechanics, which involves contractions in all directions. For example, contractility in the transverse direction has received very little attention. Contractility in the diagonal directions is no less important in whole-bladder function. Comparative studies of transverse and longitudinal contractions have been conducted by a number of groups independently [2-6]. Among these studies exists considerable variability between species chosen, age of the animals, and types of stimuli used. Any attempt to draw general conclusions on transverse and longitudinal detrusor contractility thus requires additional comparative studies.
Bladder emptying requires activation of mostly muscarinic M3 receptors, and M2 receptors to a lesser extent . Detrusor smooth muscle cells express a variety of K+ channels, including voltage-sensitive K+ (Kv) channels, ATP-sensitive K+ (KATP) channels and Ca2+-activated K+ channels of different conductances, ie, large-conductance (BK), intermediate-conductance (IK), and small-conductance (SK) [7,8]. In addition to regulating myogenic phasic contractions [9,10], these channels may also modulate drug-stimulated detrusor contractions differently in transverse and longitudinal directions. Hence, concentration-varied effects of Kv, KATP, BK, IK, and SK channel blockers on subsequent carbachol (CCh)-induced rat detrusor contractions in both transverse and longitudinal strips were examined in this study. Findings would provide information concerning K+ channel-modulated heterogeneity of rat detrusor contractions in the transverse and longitudinal directions. Recognizing the different contractile properties according to tissue orientations will have implications in making generalizations about contractile data obtained from only 1 direction, which is most conventionally longitudinal.
All procedures were performed according to rules outlined by the Institutional Animal Care and Use Committee at Nanyang Technological University, Singapore (Project approval No. ARF SBS/NIE-A 003). Six- to 7-week-old Sprague-Dawley rats of either gender were killed by CO2 asphyxiation. The whole bladder was excised as previously described  and immediately placed in carbogen-aerated ice-cold Krebs' solution. The bladder base, which made up about one-third of the bladder, was discarded. Tissues isolated from both the dorsal and ventral sides of the bladder dome were used. The bladder dome was opened to expose the urothelium by cutting along both longitudinal sides. Fine pins were used to fix the tissue on a Sylgard® (Dow Corning Corp, Singapore) coated petri dish. Using a razor blade, 2 transverse and 2 longitudinal detrusor strips (5 mm Ã—1 mm) with the urothelium intact were dissected. The strips were mounted on a tissue myograph system (Danish Myo Technology Model 800MS, Denmark) containing Krebs' solution at 37ºC. Isometric tension was monitored in both transverse and longitudinal directions and recorded using a Powerlab interface and the LabChart software (ADInstruments, Australia).
The detrusor strips were allowed to equilibrate for 30 minutes with multiple washouts at 2 g of resting tension. The viability of the strips was then tested using K+-Krebs' solution bubbled with a mixture of 95% oxygen and 5% carbon dioxide. No difference in responses to K+-Krebs' solution was observed between transverse and longitudinal strips, as reported by others [3,4,6]. After an additional 30 minutes of continuous washout, cumulative concentration-response curves (CRCs) of the muscarinic agonist CCh were constructed with and without prior treatment of a K+ channel blocker. In between each CCh CRC, 30 minutes of continuous washout with Krebs' solution was applied. A low, intermediate, and high concentration of each K+ channel blocker was used to treat the tissues for 15 minutes. The K+ channel blockers used in this study were tetraethylammonium chloride (TEA), 4-aminopyridine (4-AP), glibenclamide (Glib), iberiotoxin (IbTX), charybdotoxin (ChTX) and apamin. Time-matched controls of cumulative CCh contractions were done to ensure that tissues were not desensitized after repeated exposures to CCh. Data for each K+ channel blocker treatment were collected from 10 transverse and 10 longitudinal strips from 5 rat bladders.
Drugs and Chemicals
The composition of Krebs' solution was as follows (in mM): NaCl (119), MgCl2 (1.2), NaH2PO4 (1.2), NaHCO3 (15), KCl (4.6), CaCl2 (1.5), Glucose (11). For K+-Krebs' solution, no NaCl was added; 124 mM KCl was used instead. All other constituents remained the same. All chemicals and drugs used in this study were purchased from Sigma-Aldrich Co. (Singapore). Unless specified, all drugs were dissolved in Ca2+-free Krebs' solution. Glib was dissolved in dimethyl sulfonide; 4-AP was dissolved in 70% ethanol; apamin was dissolved in 0.05 M acetic acid. For drugs dissolved in solvents other than Ca2+-free Krebs' solution, a minimum of 1-to-1000 dilution from the stock drug solution was performed to prevent nonspecific effects elicited on the tissues.
Calculations of the CRC data and statistical analysis were done using the Prism 4 software (GraphPad Software Inc, USA). The peak response after addition of the K+-Krebs' solution was considered as 100% contraction, to which CCh CRCs were normalized. In consideration of the phasic contractions sometimes observed, an average value was obtained over a 10-second interval (from LabChart) after each CCh addition had reached a steady-state. The average value was used to plot the CRCs.
A two-way ANOVA and Bonferroni post-hoc test were used to compare amplitudes of varying concentrations of CCh-induced contractions across different treatments of the same K+ channel blocker. The CRC data were also fitted by nonlinear regression to determine the potency of CCh (expressed as log EC50) and the maximal efficacy of CCh (expressed as Emax). A one-way ANOVA and Bonferroni post-hoc test were used to compare log EC50 values. All data shown in graphs and tables were mean values ± standard error of the mean (SEM). Probability values < .05 were considered statistically significant.
Response to Carbachol After Tetraethylammonium Chloride Pretreatment
TEA is a nonselective agent with inhibitory effects on Kv, BK, and KATP channels. In both transverse and longitudinal strips, significant effects on CCh contraction were observed only when treated with 10-2 M TEA. In transverse strips, responses to 3Ã—10-7 M CCh were smaller than control in the presence of 10-2 M TEA (Figure 1). In longitudinal strips, the same TEA treatment resulted in smaller responses to CCh over a wider concentration range, from 3Ã—10-7 to 3Ã—10-6 M. The potency of CCh, expressed as log EC50, was significantly lower in longitudinal strips after 10-2 M TEA treatment Table 1. No change in CCh potency was found in transverse strips. The maximal fitted responses (Emax) to the highest CCh concentration (ie, 3Ã—10-5 M CCh) were not altered by the TEA treatments (Table 1). The findings from nonselective blockade by TEA suggested composite effects of K+ channels in mediating CCh-induced transverse and longitudinal contractions to varying degrees. Experiments with selective K+ channel blockers were subsequently conducted to reveal differences in transverse and longitudinal contractile responsiveness.
Response to Carbachol After 4-aminopyridine Pretreatment
4-AP was used to block Kv channels. In the presence of 4-AP (10-4 to 10-2 M), the CCh CRCs for both transverse and longitudinal strips were shifted to the right and suppressed (Figure 2). The potency of CCh was significantly lower in both transverse and longitudinal strips after 10-3 and 10-2 M 4-AP treatments Table 1. The Emax to 3Ã—10-5 M CCh were significantly lower in both transverse and longitudinal strips under 10-2 M 4-AP treatment. The overall effects of 4-AP in CCh-induced contractions were indistinguishable between transverse and longitudinal strips.
Response to Carbachol After Glibenclamide Pretreatment
ATP-sensitive K+ (KATP) channels were blocked by Glib. Although the KATP blockade decreased overall CCh-induced contractility in both transverse and longitudinal strips, the relationship between Glib concentrations and the affected ranges of CCh concentrations was less clear. In transverse strips, sensitivity to 10-9 M Glib was higher than to other Glib concentrations. Both 10-6 and 3Ã—10-6 M CCh-induced contractions were smaller under 10-9 M Glib treatment, whereas 10-8 M Glib failed to significantly suppress responses to all CCh concentrations (Figure 3). In longitudinal strips, however, the suppressant effect of 10-9 and 10-8 M Glib on CCh responses was not different. In the presence of 10-7 M Glib, both transverse and longitudinal strips showed similarly diminished CCh responses. The potency of CCh was not altered by Glib in either transverse or longitudinal strips, but longitudinal contractions may be more sensitive to higher concentrations of Glib, as indicated by the significantly smaller Emax Table 1. In summary, inhibitory effects on CCh-induced contractions were more pronounced in the longitudinal direction, especially when Glib concentrations were higher (ie, above 10-9 M).
Response to Carbachol After Iberiotoxin, Charybdotoxin, and Apamin Pretreatment
Ca2+-activated K+ channels of different conductance were blocked by the following: IbTX (10-9 to 10-7 M) for BK, ChTX (10-9 to 10-7 M) for both BK and IK, and apamin (10-9 to 10-7 M) for SK. In both transverse and longitudinal strips, IbTX had no significant effect on the contractile responses, including potency of CCh (Figure 4, Table 1). The same was true for CCh potency for ChTX, but blocking BK and IK channels revealed dual effects depending on the CCh concentration. In both transverse and longitudinal strips, contractions to 10-7 and 3Ã—10-7 M CCh were larger under 10-7 M ChTX treatment; statistical significance was achieved in longitudinal strips at 3Ã—10-7 M CCh (Figure 5). Other ChTX concentrations (ie, 10-9 and 10-8 M) failed to elicit this potentiating effect on CCh-induced contractions. Instead, the lower ChTX concentrations suppressed responses to CCh to varying extent, depending on the contractile direction. In transverse strips, CCh contractions were significantly diminished at 3Ã—10-6 M only (Figure 5). In longitudinal strips, significantly smaller contractions were observed over the range of 3Ã—10-6 to 3Ã—10-5 M CCh. The results from ChTX treatment suggested that longitudinal contractions induced by CCh were more sensitive to combined BK and IK channel blockade.
All concentrations of apamin suppressed responses to CCh at 10-6 M and above in transverse strips (Figure 6). Potency of CCh was also lower under 10-8 and 10-7 M apamin treatments in transverse strips. In longitudinal strips, 10-9 M apamin failed to affect CCh-induced contractions (Figure 6). On the other hand, higher concentrations of apamin (i.e. 10-8 and 10-7M) suppressed the responses to CCh at 3Ã—10-6 M and above in longitudinal strips. At high CCh concentrations, the inhibitory effect of apamin in longitudinal strips was more pronounced than in transverse strips. Specifically, the fitted maximal response to CCh was significantly diminished by both 10-8 and 10-7 M apamin in longitudinal strips, but potency of CCh remained unchanged Table 1. A distinction between transverse and longitudinal contractile responses was therefore revealed by apamin treatment. Maximal contractile responses to CCh in the longitudinal direction were more sensitive to SK channel blockade. However, potency of CCh was subject to greater changes when contractions were measured in the transverse direction.
Many aspects of detrusor contractility have been extensively investigated over the years. A vast number of studies used detrusor strips that were isolated longitudinally from bladders of different species. In contrast, only a handful of studies examined the contractility of transverse detrusor strips or compared them with longitudinal strips. The very limited information in the literature added to the difficulty in reaching sound conclusions about the contractile properties of the detrusor based on tissue orientations. Much of the variability in the findings could be attributed to the species chosen or the age of the animals, as exemplified by different studies on phasic contractions [2,4,6]. Moreover, the type of stimuli or pharmacological treatment also contributes to the heterogeneity between transverse and longitudinal strips . The overall orientation of smooth muscle cells in the rat detrusor is still under debate [12,13]. In spite of known histological differences between human and rat detrusor, the latter is still extensively used in studying bladder function. Rat models with functional bladder abnormalities and disease-associated secondary bladder dysfunction are well established [14,15]. It is therefore justifiable to continue using the rat bladder as a tool in detailing directional contractile differences, and to extend these studies to examine diseased bladders in the future. Such research may provide useful ideas related to changes in bladder function due to aging or disease.
In the present study, young adult rat detrusor contractility in transverse and longitudinal directions was monitored under the influence of different K+ channel blockers. The putative roles played by various K+ channels in mediating bladder contractions was considered . In particular, the functional activity of K+ channels during active contraction (ie, urine release) was manipulated by pretreatment with K+ channel blockers followed by CCh stimulation. The K+ channel blockers chosen here selectively inhibit the activity of Kv, KATP, BK, and SK channels [16,17]. In agreement with findings by others [3,4], no intrinsic contractile difference to stimulation with a muscarinic agonist was observed between transverse and longitudinal strips. In this study, varied results were obtained concerning transverse or longitudinal contractile sensitivity to the blockade of different subtypes of K+ channels.
Three interrelated parameters were used to assess the contractile responsiveness of transverse and longitudinal detrusor strips to CCh stimulation: (1) potency of CCh was expressed as log EC50 values; (2) efficacy of CCh was compared at each CCh concentration applied; (3) efficacy of CCh was determined at the maximal fitted CCh response (see Methods). The more general effects of K+ channel blockade were first examined with the nonselective blocker TEA. A stronger effect on longitudinal contractions was demonstrated by decreased CCh potency and suppressed responses, when compared with those of transverse strips Table 1, Table 1. The observed differences in TEA-treated strips suggested that modulation of K+ channel activity could alter direction-specific detrusor contractility. Because of its indiscriminate blockade of most K+ channel subtypes, findings from the TEA experiments led to 2 unresolved questions which are subsequently defined.
Do All K+ Channels Contribute a Greater Share to CCh-induced Contractions in 1 Direction or Another?
This question was answered by results of the 4-AP and IbTX experiments (Figure 2, Figure 4, Table 1). Blockade of Kv channels decreased both CCh potency and efficacy, but distinction between the effects of 4-AP on transverse and longitudinal strips was not visible. Similarly, IbTX treatment did not alter CCh-induced contractions differently between transverse and longitudinal strips; there were no effects of the BK channel blockade at all. Thus, not all subtypes of K+ channels contribute to the contractile differences in transverse and longitudinal directions seen with TEA treatment.
Besides Kv and BK channels, the other subtype known to be blocked by TEA is KATP channels. Using Glib at varying concentrations, differential sensitivity of transverse and longitudinal contractility to KATP channel blockade was revealed. Figure 4 shows that 10-9 M Glib suppressed the CCh responses of both transverse and longitudinal strips. When Glib concentration was increased ten-fold, the suppressant effect on transverse strips disappeared. A yet higher concentration of Glib (10-7 M) was able to decrease CCh-induced contractility in both transverse and longitudinal directions, but the effect on the latter was more pronounced. These findings suggest that compared to transverse strips, longitudinal strips were increasingly sensitive to KATP channel blockade when Glib concentration was higher. The stronger effect of KATP channel blockade on longitudinal contractions agreed incidentally with the results of the TEA experiment after Kv and BK effects were ruled out.
As mentioned above, the selective BK channel blocker IbTX did not alter CCh contractions in both transverse and longitudinal strips. The effects of IK channel blockade could therefore be isolated by treatment with ChTX, which inhibits both BK and IK channel activity. In general, longitudinal contractions were more sensitive to IK channel blockade. Responses to low concentrations of CCh were increased under ChTX treatment, but the opposite was true when CCh concentrations were higher Figure 5. This may indicate augmented contractility from additional depolarization as a result of inhibited K+ efflux via IK channels, occurring only at a low excited state (ie, at low CCh concentrations).
Contractile differences between transverse and longitudinal strips were most evident under SK channel blockade by apamin. Exposure to a low concentration of apamin (10-9 M) suppressed transverse contractions to submaximal CCh concentrations, but did not alter longitudinal contractility at all Figure 6. Higher apamin concentrations reduced transverse contractility further, but the inhibitory effect was more pronounced in longitudinal strips. Specifically, longitudinal contractility was diminished the most at higher CCh concentrations, exemplified by the significantly smaller Emax Table 1. Thus, transverse and longitudinal strips responded with different sensitivity to SK channel blockade, depending on the concentration of apamin. This was similar to the phenomenon observed with Glib treatment. Although apamin had greater effects on longitudinal contractility, CCh potency was not altered Table 1. Instead, TR strips showed significantly lower potency to CCh, despite being less sensitive to apamin-modulated contractility. Thus, the extent of the contracted state (ie, low vs high CCh concentrations), may be a crucial predetermining factor in the contractile direction(s) to be studied and the measuring parameter(s) to be analyzed.
Is the Transverse or Longitudinal Direction Consistently More Responsive to K+ Channel Blockade?
A primary aim of the present study was to identify and highlight differences in the direction of detrusor contractions. Building on earlier reports by others on active muscarinic-stimulated contractions, the results revealed variability among K+ channel subtypes in mediating transverse and longitudinal contractility. However, the lack of convincing histological data remains a limitation when interpreting contractile data for detrusor smooth muscle cells that are oriented in transverse or longitudinal directions. It has been reported that there is prevalent distribution of transverse and longitudinal smooth muscle cells in either the dorsal or ventral face of the rat bladder, but the results are conflicting [12,13]. Thus, the dorsal and ventral origin of the detrusor strips were not distinguished in the present experiments. Under control conditions (ie, in the absence of K+ channel blockade), variability was minimal between strips contracting in transverse and longitudinal directions. This suggests that all smooth muscle cells, regardless of their orientation and anatomical location, respond similarly in physiological conditions. It was after K+ channel blockade when directional contractile heterogeneity was displayed. The varying responses elicited by different K+ channel blockers may implicate functional significances of the respective channels in regulating bladder contractions in different directions. It is probable that one direction of contraction may become dysfunctional in early disease development, due to its greater dependence on a specific subtype of K+ channels. The altered contractility may translate to ineffective urine storage or release in vivo. Therefore, drug-induced detrusor contractility should preferably be examined both transversely and longitudinally to avoid overlooking any directional differences that may be present. Based on the findings presented here, more specialized techniques may also be developed in the future to isolate individual smooth muscle cells based on their orientations and to quantitate their relative prevalence in generating tension. Consequently, bladder function as a coordinated effort by individual smooth muscle cells that are oriented differently can be understood in greater detail.
Varied results were obtained concerning transverse or longitudinal contractile sensitivity to blockade of KATP, IK, and SK channels. In contrast, inhibition of Kv and BK channel activity did not reveal any underlying differences between transverse and longitudinal contractility. Contractile responses to CCh were affected to different extents in transverse and longitudinal strips depending on the type of channel blockade, the concentration of the blocker used, and the extent of the contracted state. These findings could set the stage for future investigations in the modulatory roles of other ion channels in pharmacologically-mediated contractility, not only in isolated strips but also in the whole bladder. Furthermore, technological advances may one day allow for the functional study of individual smooth muscle cells according to their orientations in the bladder.
This work was supported by Nanyang Technological University Start-Up Grant (SUG 15/07).
Conflict of Interest: none declared
- Longhurst PA, Uvelius B. Pharmacological techniques for the in vitro study of the urinary bladder. J Pharmacol Toxicol Methods. 2001;45(2):91-108. PubMed; Crossref
- Potjer RM, Constantinou CE. Frequency of spontaneous contractions in longitudinal and transverse bladder strips. Am J Physiol. 1989;257(4 Pt 2):R781-R787. PubMed
- Pagala MK, Tetsoti L, Nagpal D, Wise GJ. Aging effects on contractility of longitudinal and circular detrusor and trigone of rat bladder. J Urol. 2001;166(2):721-727. PubMed; Crossref
- Pagala M, Lehman DS, Morgan MP, Jedwab J, Wise GJ. Physiological fatigue of smooth muscle contractions in rat urinary bladder. BJU Int. 2006;97(5):1087-1093. PubMed; Crossref
- Buckner SA, Milicic I, Daza AV, Coghlan MJ, Gopalakrishnan M. Spontaneous phasic activity of the pig urinary bladder smooth muscle: characteristics and sensitivity to potassium channel modulators. Br J Pharmacol. 2002;135(3):639-648. PubMed; Crossref
- Ning Lo W, Liang W. Blockade of voltage-sensitive K+ channels increases contractility more in transverse than in longitudinal rat detrusor strips. Urology. 2009;73(2):400-404. PubMed; Crossref
- Andersson KE, Arner A. Urinary bladder contraction and relaxation: physiology and pathophysiology. Physiol Rev. 2004;84(3):935-986. PubMed; Crossref
- Thorneloe KS, Nelson MT. Properties and molecular basis of the mouse urinary bladder voltage-gated K+ current. J Physiol. 2003;549(Pt 1):65-74. PubMed; Crossref
- Brading AF. Spontaneous activity of lower urinary tract smooth muscles: correlation between ion channels and tissue function. J Physiol. 2006;570(Pt 1):13-22. PubMed; Crossref
- Herrera GM, Heppner TJ, Nelson MT. Regulation of urinary bladder smooth muscle contractions by ryanodine receptors and BK and SK channels. Am J Physiol Regul Integr Comp Physiol. 2000;279(1):R60-R68. PubMed
- Liang W, Afshar K, Stothers L, Laher I. The influence of ovariectomy and estrogen replacement on voiding patterns and detrusor muscarinic receptor affinity in the rat. Life Sci. 2002;71(3):351-362. PubMed; Crossref
- Gabella G, Uvelius B. Urinary bladder of rat: fine structure of normal and hypertrophic musculature. Cell Tissue Res. 1990;262(1):67-79. PubMed; Crossref
- Nagatomi J, Toosi KK, Grashow JS, Chancellor MB, Sacks MS. Quantification of bladder smooth muscle orientation in normal and spinal cord injured rats. Ann Biomed Eng. 2005;33(8):1078-1089. PubMed; Crossref
- McMurray G, Casey JH, Naylor AM. Animal models in urological disease and sexual dysfunction. Br J Pharmacol. 2006;147(Suppl 2):S62-S79. PubMed; Crossref
- Michel MC, Barendrecht MM. Physiological and pathological regulation of the autonomic control of urinary bladder contractility. Pharmacol Ther. 2008;117(3):297-312. PubMed; Crossref
- Hashitani H, Brading AF. Ionic basis for the regulation of spontaneous excitation in detrusor smooth muscle cells of the guinea-pig urinary bladder. Br J Pharmacol. 2003;140(1):159-169. PubMed; Crossref
- Darblade B, Behr-Roussel D, Oger S, et al. Effects of potassium channel modulators on human detrusor smooth muscle myogenic phasic contractile activity: potential therapeutic targets for overactive bladder. Urology. 2006;68(2):442-448. PubMed; Crossref