Search PMN  


PDF version
for printing

Symposium Homepage

Peer Reviewed

© 2010 Plant Management Network.
Accepted for publication 1 March 2010. Published 26 May 2010.

New Perspectives on the Epidemiology of Citrus Stubborn Disease in California Orchards

Alexandre F. S. Mello, Department of Entomology and Plant Pathology, Oklahoma State University, Stillwater, OK, 74078; Raymond K. Yokomi, San Joaquin Valley Agricultural Sciences Center, USDA-ARS, Parlier, CA, 93648; Ulrich Melcher, Department of Biochemistry and Molecular Biology, Oklahoma State University, Stillwater, OK, 74078; Jianchi Chen and Edwin Civerolo, San Joaquin Valley Agricultural Sciences Center, USDA-ARS, Parlier, CA, 93648; Astri C. Wayadande and Jacqueline Fletcher, Department of Entomology and Plant Pathology, Oklahoma State University, Stillwater, OK, 74078

Corresponding author: J. Fletcher.

Mello, A. F .S., Yokomi, R. K., Melcher, U., Chen, J., Civerolo, E., Wayadande, A. C., and Fletcher, J. 2010. New perspectives on the epidemiology of citrus stubborn disease in California orchards. Online. Plant Health Progress doi:10.1094/PHP-2010-0526-04-SY.


Although citrus stubborn disease (CSD), caused by the phloem resident mollicute Spiroplasma citri, is a significant threat to California citrus industry, our knowledge of its epidemiology is mostly anecdotal. We optimized multiple pathogen-detection protocols, measured disease incidence in two plots of commercial California groves, assessed pathogen impact on fruit quality and yield, and evaluated genetic diversity among S. citri isolates. Fruit columellas and receptacles were more suitable than leaves or bark for bacterial cultivation. Using cultivation and S. citri-specific PCR for detection, the incidence of CSD in two orchards, respectively, ranged from 46 to 85% and 1 to 4%, depending on the sampling technique. Yield and quality of fruits produced by trees that were mildly or severely CSD-symptomatic were compared to those of S. citri-free trees in one California orchard in 2006 and 2007. These infected trees had reduced fruit quality and up to 32% lower yield relative to S. citri-free trees. Using RAPD markers to compare 35 S. citri isolates collected 20 years ago from the United States and Mediterranean region with 34 isolates recently collected from California, significant genetic diversity was identified but was not correlated with the time or location of collection. Our findings suggest that CSD incidence in the commercial groves evaluated could be as high as 85% and its impact on yield and fruit quality are significant.


Citrus is a major crop in the United States, and California ranks first in fresh citrus fruit production. Historically, citrus stubborn disease (CSD) was a significant problem in the region, but between the 1980s and 1990s the severity and incidence of the disease apparently declined. However, after a series of freezes in the 1990s, the incidence of CSD in California orchards appeared to increase and the disease is once again a concern to the industry.

CSD is caused by the infection of Spiroplasma citri, a phloem-inhabiting, cell wall-less bacterium in the class Mollicutes. S. citri is transmitted by several species of leafhoppers that feed on citrus occasionally in California (11,12). Circulifer tenellus, the beet leafhopper, was reported as the major vector of the pathogen (11) but other species of leafhoppers could also be important in CSD epidemiology. S. citri can be transmitted by vectors to several weed and crop species, and the occurrence of new crop hosts, such as carrots, indicates that the host range of the vector may be increasing, and that the emergence of new vectors could be occurring (6,8).

To assure survival during environmental and host changes bacteria rely on strategies of gene evolution such as small local changes in nucleotide sequence, intragenomic reshuffling and acquisition of DNA from other organisms. Modifications of the S. citri genome were reported to occur by DNA acquisition and loss, DNA replication and repair, homologous recombination and also transposition (7). These mechanisms, alone or in combination, could generate new genes that increase the ability of S. citri to adapt to changes in the host or the environment.

CSD epidemiology is influenced by factors related to the spiroplasma, its plant hosts, vectors, the environment, and disease management practices. In California, disease incidence was variable in different locations and incidence was higher in the interior valleys compared to that in the coastal region. Transmission and symptom expression was directly related to the warmer conditions in the interior valleys (1,3). However, few epidemiological studies of stubborn disease have been performed, limiting our understanding of this complex pathosystem and our ability to develop optimal measures of management and control.

In this project, we investigated the severity and epidemiology of CSD in California. Understanding the reasons for the possible recent re-emergence of the disease, after a relatively quiescent period, will be critical in the development of management strategies that are effective, economical, and safe for people and the environment. Specific objectives of this project were to: (i) optimize sampling protocols and tools for detection of S. citri in citrus plants; (ii) analyze the incidence of stubborn disease in California orchards; (iii) assess the impact of S. citri on citrus development and production; and (iv) evaluate the genetic diversity among isolates of S. citri from different locations, countries, hosts, and time of isolation.

Optimization of Sampling Protocols and Tools for Detection of S. citri in Citrus Plants

Disease diagnosis is a key step in the monitoring and control of a pest. CSD symptoms are very similar to those of other biotic and abiotic diseases and are not reliable for determining the presence of the disease. Sweet orange [Citrus sinensis (L.) Osbeck] fruits [columella (central axis) and receptacle], flush bark, and leaves (petioles and mid-ribs) were harvested from trees in two commercial groves in Kern Co., CA, and cultured in artificial medium using standard procedures (2) (Fig. 1). Samples were incubated at 30°C and evaluated by dark-field microscopy 15 days after culturing for the presence of spiroplasmas.


Fig. 1. Center set of citrus tissues (fruit, stem, and leaves) harvested from citrus trees with citrus stubborn symptoms and on the edges, different citrus tissues that were compared to their suitability for cultivation of Spiroplasma citri.


Based on three evaluations performed in 2005 and 2006, the columella and receptacle were the most suitable tissues for detection of spiroplasmas (Table 1). Since receptacle processing required fewer filters during isolation (making the process faster and less expensive) this tissue was used in the other experiments.

Table 1. Evaluation of different citrus tissues as sources for cultivation of Spiroplasma citri. Reprinted with permission from A. F. S. Mello, et al. (10).

Tissue 1st evaluation
(Nov. 2005)
2nd evaluation
(June 2006)
3rd evaluation (Oct. 2006)
No. of positive samplesx / evaluations
Leavesy 2/6 0/7 0/11
Leaf midrib 0/6 0/7 0/11
Bark 2/6 2/7 0/11
Leaf petiole 3/6 1/7 0/11
Columella 6/6 6/7 7/11
Receptacle  NDz 6/7 7/11

 x Number of positive samples / Total number of samples.

 y Without midribs.

 z ND = not done.

Analysis of the Incidence of Stubborn Disease in California Orchards

Understanding stubborn disease epidemiology requires knowledge of the interactions among spiroplasma, plant hosts, environment, insect vectors, and management practices. Early studies in California orchards indicated that disease incidence was variable among different locations, and incidence was higher in the interior valleys than along the coast. The few stubborn epidemiology studies reported were done in the early 1970s, when molecular detection tools for mollicutes were rare and limited in scope. Therefore, an evaluation of CSD epidemiology was planned to assess the impact of the disease and its agents.

Two commercial California orchards were evaluated by three different sampling techniques. For stat sampling, every fifth tree from every fifth row was sampled and three fruit were randomly harvested per tree (Fig. 2A). In the second method, hierarchical sampling (HS), blocks of four trees were considered as a single sampling unit (Fig. 2B). Two fruit harvested from opposite sides of each tree canopy were pooled with the other fruit of the block, for a total of eight fruit per sample (5). For a third strategy, every tree blocking sampling (ETBS), six blocks of 8 by 8 trees distributed within the citrus plots were sampled. Three fruit were harvested from each of the 64 trees of each block, for a total of 384 samples per orchard. The presence of spiroplasmas was assessed by isolation from the fruit receptacle, and by PCR using primers designed from the gene for the putative adhesin, P89 (14), using DNA extracted from the columella of the same fruit used for S. citri isolation according to standard procedures (4). Disease incidences were calculated as the number of infected samples divided by the total number of samples, multiplied by 100.


Fig. 2. Field sampling techniques used to estimate citrus stubborn disease in two commercial sweet orange orchards in Kern Co., CA. (A) Stat sampling: Every fifth tree in every fifth row was sampled; each black square represents one sampled tree. (B) Hierarchical sampling (HS), each group of 4 black squares represents 4 trees pooled as a single sample. Reprinted with permission from A. F. S. Mello, et al. (10).


Significantly different CSD incidences in the two commercial citrus orchards, were demonstrated by the sampling strategies used (Table 2). Based on isolation of spiroplasma in vitro as the metric for tree infection, stat sampling indicated 45.9% disease incidence in orchard 1 and 1.3% in orchard 2 (Table 2). Hierarchical sampling, also compared to spiroplasma cultivation, indicated CSD incidences of 71.4 and 3.6%, respectively, in orchards 1 and 2. Results from the ETBS sampling (eight blocks of 64 trees) were similar to those obtained by stat sampling, yielding 50 and 1.6% CSD incidence in orchards 1 and 2, respectively.

Table 2. Incidence of citrus stubborn in two California sweet orange commercial orchards evaluated by stat, hierarchical and every-tree block sample techniques. Reprinted with permission from A. F. S. Mello, et al. (10).

Sampling method Statx Hierarchical Every-tree block
Detection method Culturing Culturing PCR Totaly Culturing PCR Totaly
    Orchard 1
Total number
of samples
74 105 105 105 382 382 382
Number of
positive samples
34  75  77  89 191 223 225
Incidence (%)z 45.9 71.4 73.3 84.8 50 58.4 58.9
    Orchard 2
Total number
of samples
78 112 112 112 377 377 377
Number of
positive samples
1 4 4 4 6 8 9
Incidence (%)z 1.3 3.6 3.6 3.6 1.6 2.1 2.4

 x Samples not evaluated by PCR

 y Number of positive samples obtained by culturing plus PCR. Positives by culturing and PCR (overlapping).

 z Number of positive samples divided by the total number of samples multiplied by 100

PCR detection of S. citri was more effective than spiroplasma isolation in vitro in the evaluation of incidence of one grove by HS and of both groves by ETBS (Table 2). Because of its superior reliability, lower cost, and rapidity, PCR of S. citri is recommended in future epidemiological studies and also in the detection of the bacteria in suspected infected plants.

Assessment of the Impact of S. citri on Citrus Development and Production

Since CSD apparently causes high impact on fruit production, we sought to estimate the impact of S. citri on fruit yield and quality. Twenty to 32 trees in one commercial orchard located in northeastern Kern Co., CA, were evaluated in 2006 and 2007. The trees were 20-year-old Thompson Improved Zimmerman Navel orange grafted onto Carrizo rootstock [C. sinensis × Poncirus trifoliata (L.) Raf]. The statistical design was based on paired trees (one healthy and one infected) and infected trees were designated mildly symptomatic ("mild") when symptoms, such as short internodes and leaf mottling, were restricted to a few stems or severely symptomatic ("severe") when the entire tree canopy was symptomatic. Confirmation of the disease status was done twice a year by culturing, PCR and q-PCR, as described by Yokomi et al. 2008 (14).

Data were collected during October from each of the 20 trees (2006) or 32 trees (2007). Evaluations included tree height and trunk circumference measurements, number of fruit dropped prematurely and estimates of yield. Thirty fruit harvested randomly from each tree were weighed and measured and the presence of sunburn and misshapen fruit (year 2007) recorded. Juice extracted from the 30 fruit was weighed, the amount of citric acid measured by a titration acidity test (TA) and the amount of soluble solids (° Brix) measured with a refractometer. The ratio between ° Brix and TA, a standard measure of citrus juice quality, was calculated.

Fig. 3. Citrus fruit randomly harvested in 2007 and used on fruit quality evaluations: fruits from S. citri-free tree (left) and fruits from S. citri-infected trees (right). S. citri confirmation was done by cultivation and PCR.


S. citri infected sweet orange trees were 13% smaller than healthy trees in 2007 (P = 0.02) but no difference in trunk circumferences were observed. Infected trees were less productive than healthy trees in both years, having fewer fruits (25 and 32% less than healthy trees in 2006 and 2007, respectively) and a higher number of fruit dropped prematurely. Severely infected trees sustained greater impact than mildly infected trees (P < 0.01). Fruit on infected trees were 13.5% lighter and smaller than those on healthy trees (Fig. 3). Significant fruit sunburn, which dries the juice vesicles, was observed only in 2007. Infected trees produced around 8% more misshapen (non-normal) fruits than did healthy trees (Fig. 4). This value became even higher (15% more) when the comparison was restricted to healthy vs. severely infected trees. No major differences between healthy and infected trees were observed in juice weight and quality evaluations.


Fig. 4. Misshaped fruit harvested from citrus trees with severe symptoms of citrus stubborn disease: fruit presenting acorn shape (left) and lopsided fruit (right).


Evaluation of the Genetic Diversity Among Isolates of S. citri from Different Locations, Countries, and Time of Isolation

Although CSD has been present in the San Joaquin Valley, CA, for many years, its impact in the region apparently increased in recent years as more growers reported disease symptoms. The S. citri genome has been shown to evolve over relatively short periods of time (13) and it is possible that the increase in CSD incidence in California orchards could be due to the occurrence of a new S. citri isolate.

Thirty five S. citri isolates obtained from 1980-1993 (designated "historic") were compared with 34 new isolates of S. citri obtained in 2005 and 2006 by cultivation from symptomatic carrots and citrus plants from several California groves (Fig. 5). Randomly amplified polymorphic DNA (RAPD-PCR) was used as the genetic marker (9). RAPD fingerprints were assessed visually and transformed into binary data (presence = 1, absence = 0). Data reliability was assessed by principal component analysis (PCA) using the SAS/PRINCOMP procedure, SAS software 9.1 (SAS institute Inc., Cary, NC).


Fig. 5. Location of citrus orchards and other sites in the San Joaquin and Antelope valleys, CA, where Spiroplasma citri was collected for this study. Each grey square represents 1.61 kg² in which a commercial orchard of 25 or more citrus trees were planted. Yellow areas on the larger map represent the main areas where citrus is grown in San Joaquin Valley. Reprinted with permission from A. F. S. Mello, et al. (9).


Transformed binary data from the RAPDs were used in a PCA to search for major clusters that would separate S. citri isolates recently cultivated from S. citri isolates obtained over 20 years ago. No major clusters separating these two groups were identified by the PCA (Fig. 6A). However, PCA of S. citri isolates cultivated from orchards 3 to 6 showed that in some orchards the genetic variability within a grove was very limited (isolates clustered very tightly) and in others the genetic variability within groves was greater (isolates were distributed in PCA) (Fig. 6B).


Fig. 6. Arrangement of Spiroplasma citri isolates based on principal component analysis using as input the differential characters obtained in random amplified polymorphism DNA reactions: (A) Analysis including historic S. citri isolates (black squares) and new isolates (open squares); (B) S. citri isolates from orchards (sites) 3 to 6. Majority of isolates from orchards 3 and 6 formed a tight cluster at principal coordinate 1 (3) and principal coordinate 2 (0). Reprinted with permission from A. F. S. Mello, et al. (9).


Significance and Importance

The enhancement of pathogen isolation efficiency and the development of molecular tools to confirm the presence of S. citri in citrus allow us to investigate stubborn-affected commercial citrus orchards at a large scale with greater reliability than was possible previously. Using these tools, we documented the current incidence of the disease in citrus orchards in the San Joaquin Valley of California, thereby justifying the increasing concerns of citrus growers.

The incidence of stubborn disease in the orchards we studied varied from low to high. Understanding the mode of inoculation is very important in the management of the disease. In the grove having low disease incidence, the infection was likely due to natural spread of the pathogen since the plot is adjacent to the foothills (natural environment of the leafhopper vector of S. citri) and the infected trees were unevenly distributed within the orchard. On the other hand, the grove having high disease incidence may have been infected through contaminated propagative material since the distribution of infected trees was dense and homogenous. Natural spread of the pathogen is very difficult to control since the vector is polyphagous and present in different locations in California; however, use of healthy propagation material is essential since reliable and sensitive techniques of detection are available (14).

In the sweet orange orchard evaluated, S. citri caused fruit impact including up to 32% yield reduction and 13% lower fruit weight. The majority of citrus produced in California goes to the fresh fruit market. In addition to the loss in productivity the increase in the number of malformed fruits increases the impact of CSD on fruit commercialization. The data obtained in this study support the recommendation that severely symptomatic trees (which had greater decrease in productivity and in fruit quality) should be removed and replaced by new plants to maintain grove productivity.

Analyses of the genetic diversity of historic and current S. citri isolates in California indicates that genome changes in the pathogen are unlikely to have led to the observed re-emergence of the disease since no major differences were identified among isolates analyzed. The total pathosystem is composed of several hosts, vectors, and environmental factors all of which play a role in disease epidemiology. Evidence of spread of the pathogen to new host species, such as carrot, demonstrate the microbe’s ability to adapt to new niches. Niche adaptability is very important since it could extend the impact of the disease to other commercial crops and prompt use of more severe measures to control the pathogen or the vector. Moreover, the existence of new plants species as sources of S. citri inoculum increases the chances of disease spread and could be a reasonable explanation for the apparent re-emergence of citrus stubborn disease in citrus orchards in California.


The authors thank G. Tanaka ( for preparing the map of sites used to sample S. citri, and C. Goad and M. E. Payton from Oklahoma State University for assistance with statistical evaluations and Asaul Gonzales, USDA-ARS Parlier, CA for assistance in harvesting fruit. This study was funded by United States Department of Agriculture-Agricultural Research Service project number 5302-22000-009-00 and by the Oklahoma Agricultural Experiment Station Hatch Project 2052.

Literature Cited

1. Bové, J. M., Calavan, E. C., Capoor, S. P., Cortez, R. E., and Schwarz, R. E. 1974. Influence of temperature on symptoms of California stubborn, South Africa greening, India citrus decline and Philippines leaf mottling diseases. Pages 12-15 in: Proc. of the 6th Conf. of the Int'l Organization of Citrus Virologists (IOCV). L. G. Weathers and M. Cohen, eds. Univ. of California, Riverside, CA.

2. Bové, J. M., Whitcomb, R. F., and McCoy, R. E. 1983. Culture techniques for spiroplasmas from plants. Pages 225-234 in: Methods in Mycoplasmology. J. G. Tully and S. Razin, eds. Academic Press Inc., New York, NY.

3. Calavan, E. C., Blue, R. L., Harjung, M. K., Cartia, G., Granett, A. L., Rana, G. L., and Gumpf, D. J. 1976. Seasonal and geographical variations in natural spread of citrus stubborn disease. Pages in: Proc. of the 7th Conf. of the Int'l Organization of Citrus Virologists (IOCV). E. C. Calavan, ed. Univ. of California, Riverside, CA.

4. Doyle, J. J., and Doyle, J. L. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem. Bull. 19:11-15.

5. Garnsey, S. M., Gottwald, T. R., and Yokomi, R. K. 1998. Control strategies for citrus tristeza virus. Pages 639-658 in: Plant Virus Disease Control. A. Hadidi, R. K. Khetarpal, and H. Koganezawa, eds. American Phytopathological Society, St. Paul, MN.

6. Lee, I.-M., Bottner, K. D., Munyaneza, J. E., Davis, R. E., Crosslin, J. M., du Toit, L. J., and Crosby, T. 2006. Carrot purple leaf: A new spiroplasmal disease associated with carrots in Washington State. Plant Dis. 90:989-993.

7. Melcher, U., and Fletcher, J. 1999. Genetic variation in Spiroplasma citri. Europ. J. Plant Pathol. 105:519-533.

8. Mello, A. F. S., Wayadande, A. C., Yokomi, R. K., and Fletcher, J. 2009. Transmission of different isolates of Spiroplasma citri to carrot and citrus by Circulifer tenellus (Hemiptera:Cicadellidae). J. Econ. Entomol. 102:1417-1422.

9. Mello, A. F. S., Yokomi, R. K., Melcher, U., Chen, J. C., Wayadande, A. C., and Fletcher, J. 2008. Genetic diversity of Spiroplasma citri strains from different regions, hosts, and isolation dates. Phytopathology 98:960-968.

10. Mello, A. F. S., Yokomi, R. K., and Fletcher, J. Assessment of citrus stubborn disease incidence in citrus. Online. Proc. of the 17th Conf. of the Int'l Organization of Citrus Virologists (IOCV), Riverside, CA.

11. Oldfield, G. N., Kaloostian, G. H., Pierce, H. D., Calavan, E. C., Granett, A. L., and Blue, R. L. 1976. Beet leafhopper transmits citrus stubborn disease. Cal. Agric. 30:15.

12. Oldfield, G. N., Kaloostian, G. H., Pierce, H. D., Calavan, E. C., Granett, A. L., Blue, R. L., Rana, G. L., and Gumpf, D. J. 1977. Transmission of Spiroplasma citri from citrus to citrus by Scaphytopius nitridus. Phytopathology 67:763-765.

13. Ye, F., Melcher, U., Rascoe, J. E., and Fletcher, J. 1996. Extensive chromosome aberrations in Spiroplasma citri strain BR3. Biochemical Genet. 34:269-286.

14. Yokomi, R. K., Mello, A. F. S., Saponari, M., and Fletcher, J. 2008. Polymerase chain reaction-based detection of Spiroplasma citri associated with citrus stubborn disease. Plant Dis. 92:253-260.