Proboscis Extension Reflex in Apis mellifera [Honeybee] with Only One Antenna

引用 收藏 提问与回复 分享您的反馈 Cited by



Scientific Reports
Oct 2016



The proboscis extension reflex (PER) is a common classical conditioned reflex which is widely used in the neurology and ethology. In honeybees, PER experiments can train bees to associate an odor with a reward or punishment. Here we present a variation of the PER experiment in Apis mellifera that trains honeybees using only one antenna. This variation on the PER paradigm could assist research efforts in fields which study lateralization within the nervous system.

Keywords: PER (PER), Honeybee (蜜蜂), Neuroscience (神经系统科学), Odor (气味)


Vertebrate brains can solve complex tasks with regional specialization (Kandel et al., 2000) and this may also be the case in invertebrates (Frasnelli et al., 2010). To date, experimental evidence for lateralization, a form of regional specialization, in honeybees has been documented for simple olfactory classical conditioning. For example, asymmetry in use of the antennae (Letzkus et al., 2006; Rogers and Vallortigara, 2008) with associated asymmetry of anatomy (i.e., sensilla number and electroantennographic responses of antennal neurons (Anfora et al., 2010; Frasnelli et al., 2010)). The honeybee Apis mellifera displays a clear laterality in responding to learned odors (Guo et al., 2016). By training honeybees on two different versions of the well-known proboscis extension reflex (PER) paradigm (Kuwabara et al., 1957; Bitterman et al., 1983), we demonstrated that bees respond to odors better when they are trained through their right antenna. To our knowledge, that was the first report of asymmetrical learning performance in an insect. In summary, our modified protocol can be used to further explore lateralization in invertebrates.

Materials and Reagents

  1. Thin strips of tape
  2. Silicone rubber compound (Chemoment Materials; Address: building F9, room 402, 110 Kexue Road, Hefei, China. Telephone: 0551-62329017)
  3. Sponge
  4. Toothpick
  5. Plastic pipe
  6. 2.5 ml syringes
  7. 1.5 ml RNase-free tubes (Corning, Axygen®, catalog number: MCT-150-C )
  8. Bees (Apis mellifera ligustica)
  9. Ice
  10. Lemon essence (Queen Fine Foods,
  11. Vanilla essence (Queen Fine Foods,
  12. Liquid nitrogen
  13. Sodium chloride (NaCl)
  14. Sucrose
  15. 1 M sucrose solution (see Recipes)
  16. Saturated saline solution (see Recipes)


  1. Metal tube (made with copper by a local machinery plant in Nanchang, Jiangxi, China, diameter: 1 cm, length: 2 cm)
  2. Incubator (Shaoguan Thaihung Medical Instrument Company, model: LRH-250A )
  3. Small suction fan
  4. Scissors


  1. For this experiment, we use worker bees from a single drone inseminated (SDI) queen of Apis mellifera ligustica. We begin the PER experiment in the morning of the eighth day after worker bees emerge. We remove the bees from the rectangular boxes and randomly assign them into three groups: Left antenna covered (LAC), Right antenna covered (RAC), and Control. This results in two trained groups (LAC and RAC) and one untrained group (Control). We cool the bees on ice for about 5 min until they stop moving (Letzkus et al., 2006).
  2. We fix each bee in a metal tube with thin strips of tape so that only the two forelegs and head are free to move. For the two trained groups, one of their antennae is covered with a two-component silicone rubber compound to ensure that the bees are trained by a single antenna. When we cover the antennae, we place the head of the bee on a sponge. After we mix the two silicone rubber compounds, we dip a toothpick in the mixture, and then carefully cover the antennae of the bee with the mixture. In contrast, the antennae of control group bees are left free (Figure 1A). Next, all of the bees are briefly fed with one or two drops of 1 M sucrose solution (see Recipes) and then moved back to an incubator to recover.
    Note: The bees are still fixed in the metal tubes while they are inside the incubator.
  3. Two hours later, we train the bees with a positive stimulus and a negative stimulus. The bees do not need to be anesthetized at this time because they are fixed in the tube. In this experiment, lemon odor + 1 M sucrose solution is used as the positive stimulus + reward,and vanilla odor + saturated saline solution (see Recipes) is used as the negative stimulus + punishment. We mix the odor with the reward or punishment. More specifically, we add 30 µl lemon odor to 4 ml 1 M sucrose solution and 30 μl vanilla essence to 4 ml saturated saline (27.74%, w/w). The bees are trained to discriminate between these two different odors. During training, the bees are first given a stimulus, then the appropriate reward or punishment. We place a small suction fan connected with a plastic pipe behind the bees to maintain a constant flow of odor during the stimulus (Figure 2). The fan also allows us to quickly remove any residual trace of odor before beginning the training of the next bee. (Figure 1B).

    Figure 1. PER experiment. A. Antenna covered with silicone rubber compound; B. Feeding the bees with sucrose water; C. Control group bees with two antennae uncovered.

    Figure 2. The schematic diagram of how the suction fan is set up

  4. Each bee is given three trials with an interval of 6 min. Each trial contains a positive and a negative stimulus training (Masutaro, 1957).
    1. On the first trial, the bee is given a positive conditioned stimulus. A droplet is placed about 1-2 cm from the bee’s antennae using a 2.5 ml syringe with needle. The bee is removed after it has extended its proboscis and tasted the solution. If the bee does not extend its proboscis in 5 sec, we briefly touch the antennae with the stimulus drop to prompt it to give a response to the odor.
    2. Then, the bee is given the negative stimulus with the same procedure above.
    3. If a bee never extended its proboscis or was inactive, it was discarded after the first trial.
    4. After training, the bees are fed 2-3 drops of 1 M sucrose solution and returned to the incubator for overnight storage.
  5. In the morning on the next day, retention tests are performed. For the LAC and RAC group, the silicone rubber is not removed until the retention tests are finished. For the retention tests, the bees are first given the negative stimulus, and then the positive stimulus. As in the training, the stimulus droplet is placed 1-2 cm from the bee’s antennae and we wait 5 sec. If the bee extends its proboscis (within 5 sec) when it is given the positive stimulus, but not when given the negative stimulus, it is recorded as correct. If not it is recorded as wrong. Each bee is tested in three trials with an interval of 6 min. To pass the test: bees must respond correctly to the positive stimulus in at least two of the three trials and must not respond to the negative stimulus in all the three trials. After the tests are complete, the heads of the trained bees (bees that passed the test) and those of the untrained bees (control group) are cut with scissors and stored in 1.5 ml RNase-free tubes in liquid nitrogen. See Figure 3 for the whole procedure.

    Figure 3. The flow chart of the training and retention tests

Data analysis

The data of our experiment has been published in Scientific Reports (Guo et al., 2016). The training results are shown in Table 1.

Table 1. PER training result


When bees are given saturated saline solution as negative stimulus, it is important to only let them eat a little saturated saline solution. We have noticed that bees that consume a lot of the saturated saline solution tend to die after training.


  1. 1 M sucrose solution
    342 g sucrose
    1,000 ml distilled water
  2. Saturated saline solution
    36.5 g NaCl
    100 ml distilled water
    Sterilized at 121 °C for 20 min


We acknowledge Prof. Yong Cao for providing the silicon compound for covering the antenna. This work was supported by Chinese Academy of Science Strategic Project of Leading Science and Technology (XDA01020402), the National High Technology Research and Development Program (“863” Program) of China (2012AA020402 and 2012AA02A202), the Earmarked Fund for China Agriculture Research System (No. CARS-45-KXJ12), the National Natural Science Foundation of China (No. 31260524) and the Research Fund for the Doctoral Program of Higher Education of China (No. 20123603120005). The authors declare that there are no conflicts of interest or competing interest.


  1. Anfora, G., Frasnelli, E., Maccagnani, B., Rogers, L. J. and Vallortigara, G. (2010). Behavioural and electrophysiological lateralization in a social (Apis mellifera) and in a non-social (Osmia cornuta) species of bee. Behav Brain Res 206: 236-239.
  2. Bitterman, M. E., Menzel, R., Fietz, A. and Schafer, S. (1983). Classical conditioning of proboscis extension in honeybees (Apis mellifera). J Comp Psychol 97(2): 107-19.
  3. Frasnelli, E., Anfora, G., Trona, F., Tessarolo, F. and Vallortigara, G. (2010). Morpho-functional asymmetry of the olfactory receptors of the honeybee (Apis mellifera). Behav Brain Res 209: 221-225.
  4. Guo, Y., Wang, Z., Li, Y., Wei, G., Yuan, J., Sun, Y., Wang, H., Qin, Q., Zeng, Z., Zhang, S. and Chen, R. (2016). Lateralization of gene expression in the honeybee brain during olfactory learning. Sci Rep 6: 34727.
  5. Letzkus, P., Ribi, W. A., Wood, J. T., Zhu, H., Zhang, S. W. and Srinivasan, M. V. (2006). Lateralization of olfaction in the honeybee Apis mellifera. Curr Biol 16(14): 1471-1476.
  6. Masutaro, K. (1957). Bildung eines bedingten Reflexes von PAVLOVs Typus bei der Honigbiene, Apis mellifica. J Fac Sci Hokkaido Univ Ser VI Zool 13(1-4): 458-464.
  7. Kandel, E. R., Schwartz, J. H. and Jessell, T. M. (2000). Principles of Neural Science. McGraw Hill.
  8. Kuwabara, M. (1957). Bildung des bedingten Reflexes von Pavlovs Typus bei der Honigbiene, Apis mellifica. J Fac Sci 13: 458-464.
  9. Rogers, L. J. and Vallortigara, G. (2008). From antenna to antenna: lateral shift of olfactory memory in honeybees. PLoS One 3(6): e2340.


长鼻延长反射(PER)是一种常见的经典条件反射,广泛应用于神经病学和行为学。 在蜜蜂,PER实验可以训练蜜蜂把气味与奖励或惩罚联系起来。 在这里,我们展示了一个只使用一个天线来培养蜜蜂的“爱荷华蜜蜂”PER实验的变种。 PER模式的这种变化可能有助于在研究神经系统内侧化的领域进行研究。

【背景】脊椎动物的大脑可以解决复杂的任务与区域专业化(坎德尔等人,2000年),这也可能在无脊椎动物的情况下(Frasnelli等人,2010年)。迄今为止,关于蜜蜂的区域专业化的侧面化的实验证据已经记载为简单的嗅觉经典调理。例如,使用天线(Letzkus等人,2006; Rogers和Vallortigara,2008)的不对称性与相关的解剖学不对称性(即感觉,感觉数量和电子天线响应的触角神经元(Anfora等人,2010; Frasnelli等人,2010))。蜜蜂(Apis mellifera)在响应学习气味方面表现出明显的偏侧性(Guo et al。,2016)。通过对熟知的长鼻延伸反射(PER)范式(Kuwabara等人,1957; Bitterman等人,1983)的两种不同版本的蜜蜂进行培训,我们证明,当蜜蜂通过正确的天线进行训练时,蜜蜂对气味的反应更好。据我们所知,这是昆虫不对称学习表现的第一个报告。总之,我们修改后的协议可以用来进一步探索无脊椎动物的侧向化。

关键字:PER, 蜜蜂, 神经系统科学, 气味


  1. 薄胶带
  2. 硅橡胶复合物(化学试剂;地址:中国合肥科学路110号402室F9室电话:0551-62329017)
  3. 海绵
  4. 牙签
  5. 塑料管
  6. 2.5毫升注射器
  7. 1.5毫升无RNA酶的试管(Corning,Axygen®,产品目录号:MCT-150-C)
  8. 蜜蜂( Apis mellifera ligustica)

  9. 柠檬精华(Queen Fine Foods, /天然柠檬萃取物-50ml /
  10. 香草精华(Queen Fine Foods, / natural-vanilla-extract-200ml /
  11. 液氮
  12. 氯化钠(NaCl)
  13. 蔗糖
  14. 1 M蔗糖溶液(见食谱)
  15. 饱和盐溶液(见食谱)


  1. 金属管(由江西南昌的一家当地机械厂制造,直径1厘米,长度2厘米,由铜制成)
  2. 孵化器(韶关泰兴医疗器械公司,型号:LRH-250A)
  3. 小吸风扇
  4. 剪刀


  1. 在这个实验中,我们使用了来自一只来自Apis mellifera ligustica的单一无人授精(SDI)女王的工蜂。工蜂出现后的第八天早上我们开始PER实验。我们从矩形框中移除蜜蜂,随机分配到三组:左侧天线覆盖(LAC),右侧天线覆盖(RAC)和控制。这导致两个训练组(LAC和RAC)和一个未训练组(对照)。我们在蜜蜂冰上冷却约5分钟,直到它们停止移动(Letzkus et。,2006年)。
  2. 我们把每个蜜蜂都固定在一个带有细条的金属管里,这样只有两个前腿和头部可以自由移动。对于两个训练组,他们的天线之一覆盖有双组分硅橡胶化合物,以确保蜜蜂由一个天线训练。当我们覆盖触角时,我们把蜂头放在海绵上。在混合两种硅橡胶混合物之后,我们在混合物中蘸一根牙签,然后用混合物小心地覆盖蜜蜂的天线。相反,对照组蜜蜂的天线是空闲的(图1A)。接下来,所有的蜜蜂都简单地喂一两滴1M蔗糖溶液(见食谱),然后移回孵化器恢复。
  3. 两个小时后,我们用积极的刺激和消极的刺激来训练蜜蜂。蜜蜂现在不需要麻醉,因为它们固定在管中。在这个实验中,柠檬气味+ 1M蔗糖溶液被用作正刺激+奖励,而香草气味+饱和盐水溶液(见食谱)被用作负刺激+惩罚。我们把气味与奖励或惩罚混合在一起。更具体地说,我们添加30微升柠檬气味4毫升1M蔗糖溶液和30微升香草精华4毫升饱和盐水(27.74%,W / W)。蜜蜂被训练来区分这两种不同的气味。在训练过程中,蜜蜂首先得到刺激,然后进行适当的奖励或惩罚。我们在蜜蜂后面放置一个与塑料管连接的小型吸风扇,以在刺激过程中保持恒定的气味流动(图2)。风扇还可以让我们在开始下一个蜜蜂训练之前迅速清除残留的痕迹。 (图1B)。

    图1. PER实验。 :一种。硅橡胶复合天线覆盖; B.用蔗糖水喂养蜜蜂; C.两只天线被发现的对照组蜜蜂。


  4. 蜜蜂每隔6分钟进行三次试验。每个审判包含积极和消极的刺激训练(Masutaro,1957)。
    1. 在第一次试验中,蜜蜂获得积极的条件刺激。使用2.5毫升针头注射器从蜜蜂的触角放置约1-2厘米的液滴。蜜蜂伸出长鼻后,将其取出,并尝试解决方案。如果蜜蜂在5秒内没有伸出它的长鼻,我们用刺激下降的方式短暂触摸天线,以提示它对气味做出反应。
    2. 然后,蜜蜂被给予与上述相同的程序的负面刺激。
    3. 如果一只蜜蜂从来没有延长它的长鼻,或者没有活动,它在第一次审判后就被丢弃了。
    4. 培训结束后,蜜蜂喂2-3滴1M蔗糖溶液,并返回孵化器过夜储存。
  5. 在第二天早上,进行保留测试。对于LAC和RAC组,在保留测试完成之前,硅橡胶不会被除去。对于保留测试,蜜蜂首先被给予负面刺激,然后是正面刺激。就像在训练中一样,刺激液滴放置在离蜜蜂触角1-2厘米处,我们等待5秒钟。如果蜜蜂在长时间(5秒内)伸出它的长鼻(当它被给予正刺激时),但当给予负刺激时没有伸长,它被记录为正确的。如果不是,则记录为错误。每只蜜蜂都以6分钟的间隔进行三次试验。为了通过测试:蜜蜂必须在三次试验中的至少两次中正确地对正面刺激作出反应,并且不得在所有三次试验中对负面刺激作出反应。测试结束后,将经过训练的蜜蜂(通过测试的蜜蜂)和未经训练的蜜蜂(对照组)的头部用剪刀剪下,并储存在1.5ml无RNase的液氮管中。



我们的实验数据已经发表在“科学报告”中(Guo et al。,2016)。培训结果见表1.

表1. PER培训结果




  1. 1 M蔗糖溶液
  2. 饱和盐水溶液



我们承认曹教授提供硅化合物覆盖天线。这项工作得到了中国科学院科技攻关计划(XDA01020402),国家高技术研究发展计划(“863”计划)(2012AA020402和2012AA02A202),中国农业科学研究院系统专项资金(No. CARS-45-KXJ12),国家自然科学基金(No. 31260524)和中国高等学校博士学科点专项(20123603120005)。作者声明不存在利益冲突或利益冲突。


  1. Anfora,G.,Frasnelli,E.,Maccagnani,B.,Rogers,L.J。和Vallortigara,G。(2010)。 行为学和电生理学侧面化在一个社会( Apis mellifera )和在一个非社会( Osmia cornuta )蜜蜂的物种。行为脑解决方案206:236-239。
  2. Bitterman,M.E.,Menzel,R.,Fietz,A。和Schafer,S。(1983)。 经典的蜜蜂喙伸展调理( Apis mellifera )。 J Comp Psychol 97(2):107-19。
  3. Frasnelli,E.,Anfora,G.,Trona,F.,Tessarolo,F.和Vallortigara,G.(2010)。 蜜蜂嗅觉受体的形态功能不对称( Apis mellifera )。 Behav Brain Res 209:221-225。
  4. 国家自然科学基金面上项目资助项目,国家自然科学基金面上项目,国家自然科学基金面上项目,国家自然科学基金面上项目陈,R.(2016)。 嗅觉学习期间蜜蜂脑中基因表达的横向化 Sci Rep 6:34727。
  5. Letzkus,P.,Ribi,W.A.,Wood,J.T.,Zhu,H.,Zhang,S.W。和Srinivasan,M.V。(2006)。 蜜蜂嗅觉横向化 Apis mellifera Curr Biol 16(14):1471-1476。
  6. Masutaro,K.(1957)。 Bildung eines bedingten Reflexes von PAVLOVs Typus bei der Honigbiene,Apis mellifica。 Sc北海道大学六年级Zool 13(1-4):458-464。
  7. Kandel,E.R。,Schwartz,J.H。和Jessell,T.M。(2000)。神经科学原理。麦格劳•希尔。
  8. Kuwabara,M.(1957)。 影响巴甫洛夫斯山脉的香味, Apis mellifica ;。 J Fac Sci 13:458-464。
  9. Rogers,L.J。和Vallortigara,G。(2008)。 从天线到天线:蜜蜂嗅觉记忆的横向移动 PLoS一个 3(6):e2340。
  • English
  • 中文翻译
免责声明 × 为了向广大用户提供经翻译的内容, 采用人工翻译与计算机翻译结合的技术翻译了本文章。基于计算机的翻译质量再高,也不及 100% 的人工翻译的质量。为此,我们始终建议用户参考原始英文版本。 Bio-protocol., LLC对翻译版本的准确性不承担任何责任。
Copyright: © 2017 The Authors; exclusive licensee Bio-protocol LLC.
引用:Guo, Y., Wang, Z., Zeng, Z., Zhang, S. and Chen, R. (2017). Proboscis Extension Reflex in Apis mellifera [Honeybee] with Only One Antenna. Bio-protocol 7(23): e2624. DOI: 10.21769/BioProtoc.2624.